[{"configuration_id":"deepswe_1_1:aa4ad931eff45be4","board":"deepswe_1_1","model":"gpt-6-astra","variant":"mini-swe-agent - gpt-6-astra (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":74.11504424778761,"score_low":71.24964807371246,"score_high":76.98044042186275,"mean_cost_usd_per_task":6.52377356460177,"median_cost_usd_per_task":5.6717151,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-6-astra","variantLabel":"mini-swe-agent - gpt-6-astra (xhigh)","score":74.11504424778761,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":2.865396174075144,"ciPlus":2.865396174075144,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":6.52377356460177,"medianCostPerTaskUsd":5.6717151,"attempted":452,"passed":335,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_6_astra_xhigh","costBasis":"Expected launch pricing at all context lengths: $12/M uncached input, $15/M cache writes, $1.20/M cache reads, $50/M output, $2/M compute units."},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:a4e0e06e8292b616","board":"deepswe_1_1","model":"gemini-3.8-flash","variant":"mini-swe-agent - gemini-3-8-flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":73.8255033557047,"score_low":72.40819014906883,"score_high":75.24281656234058,"mean_cost_usd_per_task":2.362349413758389,"median_cost_usd_per_task":2.1098912999999997,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.8-flash","variantLabel":"mini-swe-agent - gemini-3-8-flash (high)","score":73.8255033557047,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.4173132066358676,"ciPlus":1.4173132066358818,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.362349413758389,"medianCostPerTaskUsd":2.1098912999999997,"attempted":447,"passed":330,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_8_flash_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:f029cafdaf0a4ee3","board":"deepswe_1_1","model":"claude-opus-5","variant":"mini-swe-agent - claude-opus-5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":73.64864864864865,"score_low":69.77633822227692,"score_high":77.52095907502039,"mean_cost_usd_per_task":11.837583271396396,"median_cost_usd_per_task":10.4281505,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-5","variantLabel":"mini-swe-agent - claude-opus-5 (max)","score":73.64864864864865,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":3.872310426371726,"ciPlus":3.8723104263717403,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":11.837583271396396,"medianCostPerTaskUsd":10.4281505,"attempted":444,"passed":327,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_5_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e8101649c60ecba8","board":"deepswe_1_1","model":"gpt-6-astra","variant":"mini-swe-agent - gpt-6-astra (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":73.23008849557522,"score_low":69.80659243770221,"score_high":76.65358455344823,"mean_cost_usd_per_task":5.723721956194691,"median_cost_usd_per_task":4.9622669,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-6-astra","variantLabel":"mini-swe-agent - gpt-6-astra (high)","score":73.23008849557522,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":3.4234960578730096,"ciPlus":3.4234960578730096,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":5.723721956194691,"medianCostPerTaskUsd":4.9622669,"attempted":452,"passed":331,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_6_astra_high","costBasis":"Expected launch pricing at all context lengths: $12/M uncached input, $15/M cache writes, $1.20/M cache reads, $50/M output, $2/M compute units."},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:b30a594ec5f66ce7","board":"deepswe_1_1","model":"gpt-6-astra","variant":"mini-swe-agent - gpt-6-astra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":73.23008849557522,"score_low":72.39976873936956,"score_high":74.06040825178087,"mean_cost_usd_per_task":12.369032401327434,"median_cost_usd_per_task":11.036936,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-6-astra","variantLabel":"mini-swe-agent - gpt-6-astra (max)","score":73.23008849557522,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":0.8303197562056539,"ciPlus":0.8303197562056539,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":12.369032401327434,"medianCostPerTaskUsd":11.036936,"attempted":452,"passed":331,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_6_astra_max","costBasis":"Expected launch pricing at all context lengths: $12/M uncached input, $15/M cache writes, $1.20/M cache reads, $50/M output, $2/M compute units."},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:853cb17f0e600556","board":"deepswe_1_1","model":"claude-opus-5","variant":"mini-swe-agent - claude-opus-5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":73.15436241610739,"score_low":70.09338059446893,"score_high":76.21534423774585,"mean_cost_usd_per_task":9.07220232326622,"median_cost_usd_per_task":7.928816750000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-5","variantLabel":"mini-swe-agent - claude-opus-5 (xhigh)","score":73.15436241610739,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":3.060981821638464,"ciPlus":3.060981821638464,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":9.07220232326622,"medianCostPerTaskUsd":7.928816750000001,"attempted":447,"passed":327,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_5_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e252d58fa6da2ebe","board":"deepswe_1_1","model":"claude-opus-5","variant":"mini-swe-agent - claude-opus-5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":72.82850779510022,"score_low":70.88307563794007,"score_high":74.77393995226038,"mean_cost_usd_per_task":6.076084489977728,"median_cost_usd_per_task":4.96882375,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-5","variantLabel":"mini-swe-agent - claude-opus-5 (high)","score":72.82850779510022,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.9454321571601554,"ciPlus":1.9454321571601554,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":6.076084489977728,"medianCostPerTaskUsd":4.96882375,"attempted":449,"passed":327,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_5_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:a5c5cd4b12673627","board":"deepswe_1_1","model":"gpt-6-astra","variant":"mini-swe-agent - gpt-6-astra (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":72.78761061946902,"score_low":70.19796153763716,"score_high":75.37725970130089,"mean_cost_usd_per_task":4.379527705752213,"median_cost_usd_per_task":3.7133415999999997,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-6-astra","variantLabel":"mini-swe-agent - gpt-6-astra (medium)","score":72.78761061946902,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":2.5896490818318654,"ciPlus":2.5896490818318654,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.379527705752213,"medianCostPerTaskUsd":3.7133415999999997,"attempted":452,"passed":329,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_6_astra_medium","costBasis":"Expected launch pricing at all context lengths: $12/M uncached input, $15/M cache writes, $1.20/M cache reads, $50/M output, $2/M compute units."},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:b7fe36ffe2e53058","board":"deepswe_1_1","model":"gpt-5.6-sol","variant":"mini-swe-agent - gpt-5-6-sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":72.66666666666667,"score_low":69.83684441682949,"score_high":75.49648891650385,"mean_cost_usd_per_task":8.386436346666667,"median_cost_usd_per_task":6.8414715,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-sol","variantLabel":"mini-swe-agent - gpt-5-6-sol (max)","score":72.66666666666667,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":2.8298222498371786,"ciPlus":2.8298222498371786,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":8.386436346666667,"medianCostPerTaskUsd":6.8414715,"attempted":450,"passed":327,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_sol_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:9fafa3956afdc304","board":"deepswe_1_1","model":"gemini-3.8-flash","variant":"mini-swe-agent - gemini-3-8-flash (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":71.01769911504425,"score_low":68.73689454216633,"score_high":73.29850368792218,"mean_cost_usd_per_task":1.9671024789823006,"median_cost_usd_per_task":1.7342966249999998,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.8-flash","variantLabel":"mini-swe-agent - gemini-3-8-flash (medium)","score":71.01769911504425,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":2.2808045728779263,"ciPlus":2.2808045728779263,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.9671024789823006,"medianCostPerTaskUsd":1.7342966249999998,"attempted":452,"passed":321,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_8_flash_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:20ba110314745668","board":"deepswe_1_1","model":"gpt-5.6-sol","variant":"mini-swe-agent - gpt-5-6-sol (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":70.73170731707317,"score_low":69.91259559533886,"score_high":71.55081903880748,"mean_cost_usd_per_task":4.703663860310422,"median_cost_usd_per_task":4.117039,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-sol","variantLabel":"mini-swe-agent - gpt-5-6-sol (xhigh)","score":70.73170731707317,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":0.8191117217343162,"ciPlus":0.819111721734302,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.703663860310422,"medianCostPerTaskUsd":4.117039,"attempted":451,"passed":319,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_sol_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:60791be82ce959e0","board":"deepswe_1_1","model":"claude-fable-5","variant":"mini-swe-agent - claude-fable-5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":69.91150442477876,"score_low":66.66658684930744,"score_high":73.15642200025007,"mean_cost_usd_per_task":13.414521495535714,"median_cost_usd_per_task":11.341082999999998,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-fable-5","variantLabel":"mini-swe-agent - claude-fable-5 (xhigh)","score":69.91150442477876,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":3.2449175754713195,"ciPlus":3.2449175754713195,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":13.414521495535714,"medianCostPerTaskUsd":11.341082999999998,"attempted":452,"passed":316,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_fable_5_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e5e5c5b317b1f68a","board":"deepswe_1_1","model":"claude-fable-5","variant":"mini-swe-agent - claude-fable-5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":69.72477064220183,"score_low":65.6912963913939,"score_high":73.75824489300977,"mean_cost_usd_per_task":21.634702092592594,"median_cost_usd_per_task":19.23457225,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-fable-5","variantLabel":"mini-swe-agent - claude-fable-5 (max)","score":69.72477064220183,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":4.033474250807927,"ciPlus":4.0334742508079415,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":21.634702092592594,"medianCostPerTaskUsd":19.23457225,"attempted":436,"passed":304,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_fable_5_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:1e9b21c2004faec1","board":"deepswe_1_1","model":"gpt-5.6-terra","variant":"mini-swe-agent - gpt-5-6-terra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":69.62305986696231,"score_low":67.06612091651647,"score_high":72.17999881740815,"mean_cost_usd_per_task":4.945847263858093,"median_cost_usd_per_task":4.114557,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-terra","variantLabel":"mini-swe-agent - gpt-5-6-terra (max)","score":69.62305986696231,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":2.55693895044584,"ciPlus":2.55693895044584,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.945847263858093,"medianCostPerTaskUsd":4.114557,"attempted":451,"passed":314,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_terra_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:b9f6b563ece4cecf","board":"deepswe_1_1","model":"gpt-5.6-sol","variant":"mini-swe-agent - gpt-5-6-sol (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":69.40133037694012,"score_low":67.9691851565629,"score_high":70.83347559731736,"mean_cost_usd_per_task":3.4698331108647453,"median_cost_usd_per_task":2.979421,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-sol","variantLabel":"mini-swe-agent - gpt-5-6-sol (high)","score":69.40133037694012,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.4321452203772225,"ciPlus":1.4321452203772367,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.4698331108647453,"medianCostPerTaskUsd":2.979421,"attempted":451,"passed":313,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_sol_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:dac434c8397694b1","board":"deepswe_1_1","model":"glm-5.3","variant":"mini-swe-agent - glm-5-3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":68.95787139689578,"score_low":65.93922012533709,"score_high":71.97652266845449,"mean_cost_usd_per_task":3.9933584893126386,"median_cost_usd_per_task":3.1315945999999992,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"glm-5.3","variantLabel":"mini-swe-agent - glm-5-3 (max)","score":68.95787139689578,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":3.0186512715586957,"ciPlus":3.01865127155871,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.9933584893126386,"medianCostPerTaskUsd":3.1315945999999992,"attempted":451,"passed":311,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_glm_5_3_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:df5db3bddc1856ee","board":"deepswe_1_1","model":"claude-opus-5","variant":"mini-swe-agent - claude-opus-5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":68.90380313199105,"score_low":67.73124446609795,"score_high":70.07636179788415,"mean_cost_usd_per_task":3.28978764541387,"median_cost_usd_per_task":2.4819885000000004,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-5","variantLabel":"mini-swe-agent - claude-opus-5 (medium)","score":68.90380313199105,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":1.1725586658931064,"ciPlus":1.1725586658930922,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.28978764541387,"medianCostPerTaskUsd":2.4819885000000004,"attempted":447,"passed":308,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_5_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:42fd11a99a8570ba","board":"deepswe_1_1","model":"claude-fable-5","variant":"mini-swe-agent - claude-fable-5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":68.6046511627907,"score_low":67.48380844536031,"score_high":69.7254938802211,"mean_cost_usd_per_task":9.177638214788733,"median_cost_usd_per_task":7.322686000000002,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-fable-5","variantLabel":"mini-swe-agent - claude-fable-5 (high)","score":68.6046511627907,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.120842717430392,"ciPlus":1.120842717430392,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":9.177638214788733,"medianCostPerTaskUsd":7.322686000000002,"attempted":430,"passed":295,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_fable_5_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:aefa4afd042df82c","board":"deepswe_1_1","model":"kimi-k3","variant":"mini-swe-agent - kimi-k3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":68.51441241685144,"score_low":63.97739833756913,"score_high":73.05142649613376,"mean_cost_usd_per_task":4.654682129933482,"median_cost_usd_per_task":3.6024987,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"kimi-k3","variantLabel":"mini-swe-agent - kimi-k3 (max)","score":68.51441241685144,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":4.537014079282315,"ciPlus":4.537014079282315,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.654682129933482,"medianCostPerTaskUsd":3.6024987,"attempted":451,"passed":309,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_kimi_k3_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:9517554a020a7af3","board":"deepswe_1_1","model":"grok-4.6","variant":"mini-swe-agent - grok-4-6 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":67.47787610619469,"score_low":65.19707153331677,"score_high":69.7586806790726,"mean_cost_usd_per_task":3.4489837522123894,"median_cost_usd_per_task":2.945834,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"grok-4.6","variantLabel":"mini-swe-agent - grok-4-6 (medium)","score":67.47787610619469,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":2.280804572877912,"ciPlus":2.280804572877912,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.4489837522123894,"medianCostPerTaskUsd":2.945834,"attempted":452,"passed":305,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_grok_4_6_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:7ad5b83041495e4d","board":"deepswe_1_1","model":"gpt-5.6-luna","variant":"mini-swe-agent - gpt-5-6-luna (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":67.1875,"score_low":63.19549876432895,"score_high":71.17950123567105,"mean_cost_usd_per_task":3.028116810267857,"median_cost_usd_per_task":2.2920917999999997,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-luna","variantLabel":"mini-swe-agent - gpt-5-6-luna (max)","score":67.1875,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":3.9920012356710473,"ciPlus":3.9920012356710544,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.028116810267857,"medianCostPerTaskUsd":2.2920917999999997,"attempted":448,"passed":301,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_luna_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:6d1742654e70dc7d","board":"deepswe_1_1","model":"gpt-5.5","variant":"mini-swe-agent - gpt-5-5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":67.03539823008849,"score_low":60.569751889179635,"score_high":73.50104457099735,"mean_cost_usd_per_task":7.226236674778761,"median_cost_usd_per_task":6.110482499999999,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.5","variantLabel":"mini-swe-agent - gpt-5-5 (xhigh)","score":67.03539823008849,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":6.4656463409088545,"ciPlus":6.465646340908862,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":7.226236674778761,"medianCostPerTaskUsd":6.110482499999999,"attempted":452,"passed":303,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_5_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:2d2b1fb9f6df06a3","board":"deepswe_1_1","model":"gpt-6-astra","variant":"mini-swe-agent - gpt-6-astra (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":67.03539823008849,"score_low":65.73453717840262,"score_high":68.33625928177437,"mean_cost_usd_per_task":2.188762706637168,"median_cost_usd_per_task":1.7209886,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-6-astra","variantLabel":"mini-swe-agent - gpt-6-astra (low)","score":67.03539823008849,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":1.300861051685871,"ciPlus":1.3008610516858852,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.188762706637168,"medianCostPerTaskUsd":1.7209886,"attempted":452,"passed":303,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_6_astra_low","costBasis":"Expected launch pricing at all context lengths: $12/M uncached input, $15/M cache writes, $1.20/M cache reads, $50/M output, $2/M compute units."},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:017a7f4c8bf6ada0","board":"deepswe_1_1","model":"grok-4.6","variant":"mini-swe-agent - grok-4-6 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":66.74057649667405,"score_low":64.564892272269,"score_high":68.9162607210791,"mean_cost_usd_per_task":5.497668745011087,"median_cost_usd_per_task":4.63937,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"grok-4.6","variantLabel":"mini-swe-agent - grok-4-6 (xhigh)","score":66.74057649667405,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":2.175684224405046,"ciPlus":2.175684224405046,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":5.497668745011087,"medianCostPerTaskUsd":4.63937,"attempted":451,"passed":301,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_grok_4_6_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e9d8d5305f3c7552","board":"deepswe_1_1","model":"gemini-3.7-flash","variant":"mini-swe-agent - gemini-3-7-flash (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":65.48672566371681,"score_low":62.40019338730129,"score_high":68.57325794013234,"mean_cost_usd_per_task":2.0251351826880533,"median_cost_usd_per_task":1.65422055,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.7-flash","variantLabel":"mini-swe-agent - gemini-3-7-flash (medium)","score":65.48672566371681,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":3.086532276415518,"ciPlus":3.086532276415525,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.0251351826880533,"medianCostPerTaskUsd":1.65422055,"attempted":452,"passed":296,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_7_flash_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:a1709ab0b129e07e","board":"deepswe_1_1","model":"claude-fable-5","variant":"mini-swe-agent - claude-fable-5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":65.36697247706422,"score_low":60.94504359782606,"score_high":69.78890135630239,"mean_cost_usd_per_task":6.088186528935186,"median_cost_usd_per_task":4.62221125,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-fable-5","variantLabel":"mini-swe-agent - claude-fable-5 (medium)","score":65.36697247706422,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":4.421928879238166,"ciPlus":4.421928879238166,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":6.088186528935186,"medianCostPerTaskUsd":4.62221125,"attempted":436,"passed":285,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_fable_5_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:a62d7ab2cd07619d","board":"deepswe_1_1","model":"gemini-3.7-flash","variant":"mini-swe-agent - gemini-3-7-flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":65.2654867256637,"score_low":63.47762421887939,"score_high":67.05334923244803,"mean_cost_usd_per_task":2.1762860351216813,"median_cost_usd_per_task":1.9044440625,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.7-flash","variantLabel":"mini-swe-agent - gemini-3-7-flash (high)","score":65.2654867256637,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.7878625067843146,"ciPlus":1.7878625067843217,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.1762860351216813,"medianCostPerTaskUsd":1.9044440625,"attempted":452,"passed":295,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_7_flash_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:6c1cc54c4677acba","board":"deepswe_1_1","model":"grok-4.6","variant":"mini-swe-agent - grok-4-6 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":65.18847006651885,"score_low":63.65486924786155,"score_high":66.72207088517614,"mean_cost_usd_per_task":4.3848732239467845,"median_cost_usd_per_task":3.693604,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"grok-4.6","variantLabel":"mini-swe-agent - grok-4-6 (high)","score":65.18847006651885,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.5336008186572982,"ciPlus":1.5336008186572911,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.3848732239467845,"medianCostPerTaskUsd":3.693604,"attempted":451,"passed":294,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_grok_4_6_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:8fa6331c80979ca8","board":"deepswe_1_1","model":"gpt-5.5","variant":"mini-swe-agent - gpt-5-5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":64.38053097345133,"score_low":61.26368832394585,"score_high":67.49737362295679,"mean_cost_usd_per_task":5.100437004424779,"median_cost_usd_per_task":4.582879500000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.5","variantLabel":"mini-swe-agent - gpt-5-5 (high)","score":64.38053097345133,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":3.116842649505479,"ciPlus":3.1168426495054575,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":5.100437004424779,"medianCostPerTaskUsd":4.582879500000001,"attempted":452,"passed":291,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_5_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:2bf6b5cf8d0df775","board":"deepswe_1_1","model":"glm-5.3-flash","variant":"mini-swe-agent - glm-5-3-flash (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":63.39285714285714,"score_low":59.0148863911635,"score_high":67.77082789455078,"mean_cost_usd_per_task":0.48196371245535713,"median_cost_usd_per_task":0.389975125,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"glm-5.3-flash","variantLabel":"mini-swe-agent - glm-5-3-flash (max)","score":63.39285714285714,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":4.377970751693638,"ciPlus":4.377970751693638,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.48196371245535713,"medianCostPerTaskUsd":0.389975125,"attempted":448,"passed":284,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_glm_5_3_flash_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:b317c65fa31ed642","board":"deepswe_1_1","model":"deepseek-v4-pro","variant":"mini-swe-agent - deepseek-v4-pro (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":62.83185840707964,"score_low":56.49842780985077,"score_high":69.16528900430852,"mean_cost_usd_per_task":0.24138687892256638,"median_cost_usd_per_task":0.2232531505000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"deepseek-v4-pro","variantLabel":"mini-swe-agent - deepseek-v4-pro (max)","score":62.83185840707964,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":6.333430597228876,"ciPlus":6.333430597228876,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.24138687892256638,"medianCostPerTaskUsd":0.2232531505000001,"attempted":452,"passed":284,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_deepseek_v4_pro_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:d0270ad29cebc03b","board":"deepswe_1_1","model":"gpt-5.6-sol","variant":"mini-swe-agent - gpt-5-6-sol (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":61.06194690265486,"score_low":59.47858925334765,"score_high":62.64530455196208,"mean_cost_usd_per_task":1.8620318075221238,"median_cost_usd_per_task":1.545414,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-sol","variantLabel":"mini-swe-agent - gpt-5-6-sol (medium)","score":61.06194690265486,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":1.5833576493072101,"ciPlus":1.5833576493072172,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.8620318075221238,"medianCostPerTaskUsd":1.545414,"attempted":452,"passed":276,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_sol_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:277b6c07f6ca3318","board":"deepswe_1_1","model":"gpt-5.6-terra","variant":"mini-swe-agent - gpt-5-6-terra (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":60.17699115044248,"score_low":58.052693948515355,"score_high":62.30128835236961,"mean_cost_usd_per_task":2.127191725663717,"median_cost_usd_per_task":1.8670102499999999,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-terra","variantLabel":"mini-swe-agent - gpt-5-6-terra (xhigh)","score":60.17699115044248,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":2.124297201927128,"ciPlus":2.124297201927128,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.127191725663717,"medianCostPerTaskUsd":1.8670102499999999,"attempted":452,"passed":272,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_terra_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:53874601b1d3b3da","board":"deepswe_1_1","model":"claude-fable-5","variant":"mini-swe-agent - claude-fable-5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":59.58429561200924,"score_low":56.79029016687951,"score_high":62.37830105713898,"mean_cost_usd_per_task":3.757873125874126,"median_cost_usd_per_task":2.7559204999999998,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-fable-5","variantLabel":"mini-swe-agent - claude-fable-5 (low)","score":59.58429561200924,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":2.7940054451297343,"ciPlus":2.7940054451297343,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.757873125874126,"medianCostPerTaskUsd":2.7559204999999998,"attempted":433,"passed":258,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_fable_5_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:03fa5cc9c1b4bc14","board":"deepswe_1_1","model":"claude-opus-4.8","variant":"mini-swe-agent - claude-opus-4-8 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":58.97435897435898,"score_low":57.20954362692973,"score_high":60.73917432178823,"mean_cost_usd_per_task":13.222583593240094,"median_cost_usd_per_task":12.353505999999996,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-4.8","variantLabel":"mini-swe-agent - claude-opus-4-8 (max)","score":58.97435897435898,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":1.7648153474292485,"ciPlus":1.7648153474292485,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":13.222583593240094,"medianCostPerTaskUsd":12.353505999999996,"attempted":429,"passed":253,"tasksAttempted":111,"runs":4,"config":"mini_swe_agent_claude_opus_4_8_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:3353f4eaaecc0e42","board":"deepswe_1_1","model":"claude-opus-5","variant":"mini-swe-agent - claude-opus-5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":58.12917594654788,"score_low":55.79848799986361,"score_high":60.45986389323216,"mean_cost_usd_per_task":1.6626223919821828,"median_cost_usd_per_task":1.2162672499999996,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-5","variantLabel":"mini-swe-agent - claude-opus-5 (low)","score":58.12917594654788,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":2.3306879466842716,"ciPlus":2.3306879466842787,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.6626223919821828,"medianCostPerTaskUsd":1.2162672499999996,"attempted":449,"passed":261,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_5_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:f20515f2e81cba89","board":"deepswe_1_1","model":"qwen3.8-max","variant":"mini-swe-agent - qwen3-8-max (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":57.46102449888642,"score_low":54.798231826378874,"score_high":60.12381717139396,"mean_cost_usd_per_task":3.7290646500445437,"median_cost_usd_per_task":3.132689060000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"qwen3.8-max","variantLabel":"mini-swe-agent - qwen3-8-max (xhigh)","score":57.46102449888642,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":2.6627926725075426,"ciPlus":2.6627926725075426,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.7290646500445437,"medianCostPerTaskUsd":3.132689060000001,"attempted":449,"passed":258,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_qwen3_8_max_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:7a41329b22a4d1b1","board":"deepswe_1_1","model":"gpt-5.6-luna","variant":"mini-swe-agent - gpt-5-6-luna (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":56.85840707964602,"score_low":54.69030532683623,"score_high":59.02650883245582,"mean_cost_usd_per_task":1.5356192685840708,"median_cost_usd_per_task":1.2449332,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-luna","variantLabel":"mini-swe-agent - gpt-5-6-luna (xhigh)","score":56.85840707964602,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":2.1681017528097897,"ciPlus":2.168101752809797,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.5356192685840708,"medianCostPerTaskUsd":1.2449332,"attempted":452,"passed":257,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_luna_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:c181f2262d3e9d70","board":"deepswe_1_1","model":"muse-spark-1.2","variant":"mini-swe-agent - muse-spark-1-2 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":54.86725663716814,"score_low":52.74295943524101,"score_high":56.99155383909527,"mean_cost_usd_per_task":3.6955048180309733,"median_cost_usd_per_task":2.6840506,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"muse-spark-1.2","variantLabel":"mini-swe-agent - muse-spark-1-2 (xhigh)","score":54.86725663716814,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":2.124297201927128,"ciPlus":2.124297201927128,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.6955048180309733,"medianCostPerTaskUsd":2.6840506,"attempted":452,"passed":248,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_muse_spark_1_2_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:9ba29daf22e9a1ae","board":"deepswe_1_1","model":"claude-opus-4.8","variant":"mini-swe-agent - claude-opus-4-8 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":54.36241610738255,"score_low":50.64898547928808,"score_high":58.07584673547702,"mean_cost_usd_per_task":8.006355704138702,"median_cost_usd_per_task":7.049357750000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-4.8","variantLabel":"mini-swe-agent - claude-opus-4-8 (xhigh)","score":54.36241610738255,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":3.7134306280944713,"ciPlus":3.7134306280944713,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":8.006355704138702,"medianCostPerTaskUsd":7.049357750000001,"attempted":447,"passed":243,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_4_8_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:f4c657e2ffa66931","board":"deepswe_1_1","model":"gpt-5.5","variant":"mini-swe-agent - gpt-5-5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":53.98230088495575,"score_low":51.42921338966476,"score_high":56.53538838024673,"mean_cost_usd_per_task":2.749342734513274,"median_cost_usd_per_task":2.436942,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.5","variantLabel":"mini-swe-agent - gpt-5-5 (medium)","score":53.98230088495575,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":2.5530874952909883,"ciPlus":2.5530874952909812,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.749342734513274,"medianCostPerTaskUsd":2.436942,"attempted":452,"passed":244,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_5_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:17f1ccadd79e3f73","board":"deepswe_1_1","model":"claude-sonnet-5","variant":"mini-swe-agent - claude-sonnet-5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":53.84615384615385,"score_low":49.60923737597946,"score_high":58.08307031632823,"mean_cost_usd_per_task":26.399858950791852,"median_cost_usd_per_task":23.277376725000003,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-sonnet-5","variantLabel":"mini-swe-agent - claude-sonnet-5 (max)","score":53.84615384615385,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":4.2369164701743856,"ciPlus":4.2369164701743856,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":26.399858950791852,"medianCostPerTaskUsd":23.277376725000003,"attempted":442,"passed":238,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_sonnet_5_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:c3c0d085a762f3d1","board":"deepswe_1_1","model":"gemini-3.7-flash","variant":"mini-swe-agent - gemini-3-7-flash (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":53.76106194690266,"score_low":51.171412865070785,"score_high":56.35071102873452,"mean_cost_usd_per_task":1.8322645629424779,"median_cost_usd_per_task":1.3913454375,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.7-flash","variantLabel":"mini-swe-agent - gemini-3-7-flash (low)","score":53.76106194690266,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":2.5896490818318725,"ciPlus":2.5896490818318654,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.8322645629424779,"medianCostPerTaskUsd":1.3913454375,"attempted":452,"passed":243,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_7_flash_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:31c29b4c94cf77ae","board":"deepswe_1_1","model":"gpt-5.6-terra","variant":"mini-swe-agent - gpt-5-6-terra (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":53.76106194690266,"score_low":49.432091479689106,"score_high":58.09003241411621,"mean_cost_usd_per_task":1.1343762765486725,"median_cost_usd_per_task":1.031709,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-terra","variantLabel":"mini-swe-agent - gpt-5-6-terra (high)","score":53.76106194690266,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":4.328970467213551,"ciPlus":4.328970467213551,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.1343762765486725,"medianCostPerTaskUsd":1.031709,"attempted":452,"passed":243,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_terra_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:70b6e868fcb9cd80","board":"deepswe_1_1","model":"grok-4.5","variant":"mini-swe-agent - grok-4-5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":53.76106194690266,"score_low":51.48025737402474,"score_high":56.04186651978057,"mean_cost_usd_per_task":2.4157481725663716,"median_cost_usd_per_task":2.02311,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"grok-4.5","variantLabel":"mini-swe-agent - grok-4-5 (high)","score":53.76106194690266,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":2.280804572877919,"ciPlus":2.280804572877912,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.4157481725663716,"medianCostPerTaskUsd":2.02311,"attempted":452,"passed":243,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_grok_4_5_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:c0668bd778ebe124","board":"deepswe_1_1","model":"deepseek-v4-flash","variant":"mini-swe-agent - deepseek-v4-flash (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":53.31858407079646,"score_low":49.751633855764034,"score_high":56.88553428582889,"mean_cost_usd_per_task":0.10023560370265487,"median_cost_usd_per_task":0.09124898719999996,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"deepseek-v4-flash","variantLabel":"mini-swe-agent - deepseek-v4-flash (max)","score":53.31858407079646,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":3.5669502150324277,"ciPlus":3.5669502150324277,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.10023560370265487,"medianCostPerTaskUsd":0.09124898719999996,"attempted":452,"passed":241,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_deepseek_v4_flash_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e18ab480d9cf17e0","board":"deepswe_1_1","model":"muse-spark-1.1","variant":"mini-swe-agent - muse-spark-1-1 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":53.31858407079646,"score_low":50.28324161686275,"score_high":56.35392652473017,"mean_cost_usd_per_task":2.361142644469026,"median_cost_usd_per_task":1.7171158,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"muse-spark-1.1","variantLabel":"mini-swe-agent - muse-spark-1-1 (xhigh)","score":53.31858407079646,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":3.0353424539337084,"ciPlus":3.0353424539337084,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.361142644469026,"medianCostPerTaskUsd":1.7171158,"attempted":452,"passed":241,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_muse_spark_1_1_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:2b59cc2e16364e15","board":"deepswe_1_1","model":"claude-opus-4.8","variant":"mini-swe-agent - claude-opus-4-8 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":51.76991150442478,"score_low":47.20830235866894,"score_high":56.33152065018062,"mean_cost_usd_per_task":4.281940017699115,"median_cost_usd_per_task":3.6635345,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-4.8","variantLabel":"mini-swe-agent - claude-opus-4-8 (high)","score":51.76991150442478,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":4.561609145755845,"ciPlus":4.561609145755838,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.281940017699115,"medianCostPerTaskUsd":3.6635345,"attempted":452,"passed":234,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_4_8_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:148350e563933fb6","board":"deepswe_1_1","model":"gpt-5.4","variant":"mini-swe-agent - gpt-5-4 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":51.76991150442478,"score_low":50.2678065476865,"score_high":53.27201646116306,"mean_cost_usd_per_task":5.6524577245575225,"median_cost_usd_per_task":4.3528565,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.4","variantLabel":"mini-swe-agent - gpt-5-4 (xhigh)","score":51.76991150442478,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":1.502104956738286,"ciPlus":1.502104956738279,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":5.6524577245575225,"medianCostPerTaskUsd":4.3528565,"attempted":452,"passed":234,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_4_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:6a089e055d07f388","board":"deepswe_1_1","model":"claude-sonnet-5","variant":"mini-swe-agent - claude-sonnet-5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"xhigh","service_mode":null,"score":49.667405764966745,"score_low":46.21285677328888,"score_high":53.121954756644605,"mean_cost_usd_per_task":11.890634540022173,"median_cost_usd_per_task":9.965361750000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-sonnet-5","variantLabel":"mini-swe-agent - claude-sonnet-5 (xhigh)","score":49.667405764966745,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"xhigh","ciMinus":3.4545489916778678,"ciPlus":3.4545489916778607,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":11.890634540022173,"medianCostPerTaskUsd":9.965361750000001,"attempted":451,"passed":224,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_sonnet_5_xhigh","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:7d16a6959bd957f6","board":"deepswe_1_1","model":"claude-opus-4.8","variant":"mini-swe-agent - claude-opus-4-8 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":48.67256637168141,"score_low":46.43336050994396,"score_high":50.91177223341886,"mean_cost_usd_per_task":3.4438571028761062,"median_cost_usd_per_task":2.84043425,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-4.8","variantLabel":"mini-swe-agent - claude-opus-4-8 (medium)","score":48.67256637168141,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":2.239205861737453,"ciPlus":2.239205861737446,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.4438571028761062,"medianCostPerTaskUsd":2.84043425,"attempted":452,"passed":220,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_4_8_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e833a890e9fc6ef0","board":"deepswe_1_1","model":"claude-sonnet-5","variant":"mini-swe-agent - claude-sonnet-5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":48.23008849557522,"score_low":43.72377362536038,"score_high":52.73640336579006,"mean_cost_usd_per_task":7.425560161197339,"median_cost_usd_per_task":5.833649100000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-sonnet-5","variantLabel":"mini-swe-agent - claude-sonnet-5 (high)","score":48.23008849557522,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":4.506314870214837,"ciPlus":4.506314870214844,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":7.425560161197339,"medianCostPerTaskUsd":5.833649100000001,"attempted":452,"passed":218,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_sonnet_5_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:e1743e5569cb8648","board":"deepswe_1_1","model":"gemini-3.6-flash","variant":"mini-swe-agent - gemini-3-6-flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":46.68141592920354,"score_low":42.97656202995626,"score_high":50.386269828450814,"mean_cost_usd_per_task":4.419013277333334,"median_cost_usd_per_task":3.444695325000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.6-flash","variantLabel":"mini-swe-agent - gemini-3-6-flash (high)","score":46.68141592920354,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":3.7048538992472757,"ciPlus":3.7048538992472757,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.419013277333334,"medianCostPerTaskUsd":3.444695325000001,"attempted":452,"passed":211,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_6_flash_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:eed6875ca51165fd","board":"deepswe_1_1","model":"gpt-5.6-sol","variant":"mini-swe-agent - gpt-5-6-sol (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":45.35398230088495,"score_low":42.96578762942603,"score_high":47.74217697234388,"mean_cost_usd_per_task":1.0742664646017699,"median_cost_usd_per_task":0.944557,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-sol","variantLabel":"mini-swe-agent - gpt-5-6-sol (low)","score":45.35398230088495,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":2.388194671458919,"ciPlus":2.388194671458926,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.0742664646017699,"medianCostPerTaskUsd":0.944557,"attempted":452,"passed":205,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_sol_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:47d55f64ef31558b","board":"deepswe_1_1","model":"gpt-5.6-luna","variant":"mini-swe-agent - gpt-5-6-luna (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":44.24778761061947,"score_low":41.328220362702936,"score_high":47.167354858536,"mean_cost_usd_per_task":0.7779007499999999,"median_cost_usd_per_task":0.6635542999999999,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-luna","variantLabel":"mini-swe-agent - gpt-5-6-luna (high)","score":44.24778761061947,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":2.9195672479165324,"ciPlus":2.9195672479165324,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.7779007499999999,"medianCostPerTaskUsd":0.6635542999999999,"attempted":452,"passed":200,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_luna_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:3c0de09601f2c8a8","board":"deepswe_1_1","model":"glm-5.2","variant":"mini-swe-agent - glm-5-2 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"max","service_mode":null,"score":43.77777777777778,"score_low":42.05214815719697,"score_high":45.50340739835859,"mean_cost_usd_per_task":3.9199075743111114,"median_cost_usd_per_task":3.4637479200000003,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"glm-5.2","variantLabel":"mini-swe-agent - glm-5-2 (max)","score":43.77777777777778,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"max","ciMinus":1.7256296205808113,"ciPlus":1.7256296205808113,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.9199075743111114,"medianCostPerTaskUsd":3.4637479200000003,"attempted":450,"passed":197,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_glm_5_2_max","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:baca3e04439efe09","board":"deepswe_1_1","model":"grok-4.6","variant":"mini-swe-agent - grok-4-6 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":41.648106904231625,"score_low":39.32945513980172,"score_high":43.96675866866153,"mean_cost_usd_per_task":1.0423604231625836,"median_cost_usd_per_task":0.742614,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"grok-4.6","variantLabel":"mini-swe-agent - grok-4-6 (low)","score":41.648106904231625,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":2.3186517644299016,"ciPlus":2.3186517644299016,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.0423604231625836,"medianCostPerTaskUsd":0.742614,"attempted":449,"passed":187,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_grok_4_6_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:39deebcf76251bec","board":"deepswe_1_1","model":"claude-opus-4.8","variant":"mini-swe-agent - claude-opus-4-8 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":40.79822616407982,"score_low":39.334641405510666,"score_high":42.261810922648976,"mean_cost_usd_per_task":2.293369216740577,"median_cost_usd_per_task":1.8240152500000004,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-opus-4.8","variantLabel":"mini-swe-agent - claude-opus-4-8 (low)","score":40.79822616407982,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":1.463584758569155,"ciPlus":1.463584758569155,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.293369216740577,"medianCostPerTaskUsd":1.8240152500000004,"attempted":451,"passed":184,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_opus_4_8_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:3cee9c2890b172bc","board":"deepswe_1_1","model":"claude-sonnet-5","variant":"mini-swe-agent - claude-sonnet-5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":39.77777777777778,"score_low":36.65020157144805,"score_high":42.90535398410751,"mean_cost_usd_per_task":4.079036229,"median_cost_usd_per_task":3.3573110999999995,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-sonnet-5","variantLabel":"mini-swe-agent - claude-sonnet-5 (medium)","score":39.77777777777778,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":3.127576206329728,"ciPlus":3.127576206329728,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":4.079036229,"medianCostPerTaskUsd":3.3573110999999995,"attempted":450,"passed":179,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_sonnet_5_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:71a3c9c8333b9bfd","board":"deepswe_1_1","model":"glm-5.2","variant":"mini-swe-agent - glm-5-2 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":36.283185840707965,"score_low":31.533112892786306,"score_high":41.033258788629624,"mean_cost_usd_per_task":2.8355164725663715,"median_cost_usd_per_task":2.25012998,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"glm-5.2","variantLabel":"mini-swe-agent - glm-5-2 (high)","score":36.283185840707965,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":4.750072947921659,"ciPlus":4.750072947921659,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.8355164725663715,"medianCostPerTaskUsd":2.25012998,"attempted":452,"passed":164,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_glm_5_2_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:c9919ba8b7aa7d29","board":"deepswe_1_1","model":"gemini-3.5-flash","variant":"mini-swe-agent - gemini-3-5-flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":36.06194690265487,"score_low":32.095643892134426,"score_high":40.02824991317531,"mean_cost_usd_per_task":3.4467114588691796,"median_cost_usd_per_task":2.8624087500000015,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.5-flash","variantLabel":"mini-swe-agent - gemini-3-5-flash (high)","score":36.06194690265487,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":3.9663030105204413,"ciPlus":3.9663030105204413,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":3.4467114588691796,"medianCostPerTaskUsd":2.8624087500000015,"attempted":452,"passed":163,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_5_flash_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:900c8577e3bc443d","board":"deepswe_1_1","model":"gpt-5.6-terra","variant":"mini-swe-agent - gpt-5-6-terra (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":35.11111111111111,"score_low":31.729146807704478,"score_high":38.49307541451774,"mean_cost_usd_per_task":0.5832493688888889,"median_cost_usd_per_task":0.546099,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-terra","variantLabel":"mini-swe-agent - gpt-5-6-terra (medium)","score":35.11111111111111,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":3.3819643034066296,"ciPlus":3.381964303406633,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.5832493688888889,"medianCostPerTaskUsd":0.546099,"attempted":450,"passed":158,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_terra_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:c25e237262636381","board":"deepswe_1_1","model":"kimi-k2.7-code","variant":"mini-swe-agent - kimi-k2-7-code (None)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"none","service_mode":null,"score":30.53097345132743,"score_low":30.030271799081337,"score_high":31.031675103573527,"mean_cost_usd_per_task":2.815536243274336,"median_cost_usd_per_task":2.1955079100000003,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"kimi-k2.7-code","variantLabel":"mini-swe-agent - kimi-k2-7-code (None)","score":30.53097345132743,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":null,"ciMinus":0.500701652246093,"ciPlus":0.5007016522460965,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.815536243274336,"medianCostPerTaskUsd":2.1955079100000003,"attempted":452,"passed":138,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_kimi_k2_7_code_default","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:4b4fac04da320c00","board":"deepswe_1_1","model":"claude-sonnet-5","variant":"mini-swe-agent - claude-sonnet-5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":30.51224944320713,"score_low":29.386231901764383,"score_high":31.638266984649878,"mean_cost_usd_per_task":2.1865550832962137,"median_cost_usd_per_task":1.6764955500000003,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-sonnet-5","variantLabel":"mini-swe-agent - claude-sonnet-5 (low)","score":30.51224944320713,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":1.1260175414427458,"ciPlus":1.1260175414427493,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.1865550832962137,"medianCostPerTaskUsd":1.6764955500000003,"attempted":449,"passed":137,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_sonnet_5_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:ad99a40bfe39fad2","board":"deepswe_1_1","model":"claude-sonnet-4.6","variant":"mini-swe-agent - claude-sonnet-4-6 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":29.933481152993345,"score_low":25.84133988380421,"score_high":34.02562242218249,"mean_cost_usd_per_task":5.5223517987804875,"median_cost_usd_per_task":4.867440150000001,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"claude-sonnet-4.6","variantLabel":"mini-swe-agent - claude-sonnet-4-6 (high)","score":29.933481152993345,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":4.092141269189135,"ciPlus":4.092141269189142,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":5.5223517987804875,"medianCostPerTaskUsd":4.867440150000001,"attempted":451,"passed":135,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_claude_sonnet_4_6_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:94c61fbd52e4cefa","board":"deepswe_1_1","model":"gpt-5.5","variant":"mini-swe-agent - gpt-5-5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":26.991150442477874,"score_low":24.696647220470695,"score_high":29.285653664485057,"mean_cost_usd_per_task":1.2002025442477877,"median_cost_usd_per_task":1.108106,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.5","variantLabel":"mini-swe-agent - gpt-5-5 (low)","score":26.991150442477874,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":2.2945032220071795,"ciPlus":2.294503222007183,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":1.2002025442477877,"medianCostPerTaskUsd":1.108106,"attempted":452,"passed":122,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_5_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:329cf994aaf09cfd","board":"deepswe_1_1","model":"gpt-5.6-terra","variant":"mini-swe-agent - gpt-5-6-terra (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":24.053452115812917,"score_low":23.276690748868134,"score_high":24.830213482757703,"mean_cost_usd_per_task":0.4277452605790646,"median_cost_usd_per_task":0.3989895,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-terra","variantLabel":"mini-swe-agent - gpt-5-6-terra (low)","score":24.053452115812917,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":0.7767613669447826,"ciPlus":0.7767613669447861,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.4277452605790646,"medianCostPerTaskUsd":0.3989895,"attempted":449,"passed":108,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_terra_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:d8f3d12f6e586e1d","board":"deepswe_1_1","model":"gemini-3.1-pro-preview","variant":"mini-swe-agent - gemini-3-1-pro-preview (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":11.725663716814159,"score_low":10.244568255705316,"score_high":13.206759177923006,"mean_cost_usd_per_task":2.1434045568888886,"median_cost_usd_per_task":1.7235064000000002,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gemini-3.1-pro-preview","variantLabel":"mini-swe-agent - gemini-3-1-pro-preview (high)","score":11.725663716814159,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"high","ciMinus":1.4810954611088434,"ciPlus":1.481095461108847,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":2.1434045568888886,"medianCostPerTaskUsd":1.7235064000000002,"attempted":452,"passed":53,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gemini_3_1_pro_preview_high","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:ec094e2327a62e03","board":"deepswe_1_1","model":"gpt-5.6-luna","variant":"mini-swe-agent - gpt-5-6-luna (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"medium","service_mode":null,"score":11.283185840707963,"score_low":10.452866084502313,"score_high":12.113505596913615,"mean_cost_usd_per_task":0.21630978982300886,"median_cost_usd_per_task":0.1994702,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-luna","variantLabel":"mini-swe-agent - gpt-5-6-luna (medium)","score":11.283185840707963,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"medium","ciMinus":0.8303197562056504,"ciPlus":0.8303197562056521,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.21630978982300886,"medianCostPerTaskUsd":0.1994702,"attempted":452,"passed":51,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_luna_medium","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:16412a120b9a4722","board":"deepswe_1_1","model":"gpt-5.6-luna","variant":"mini-swe-agent - gpt-5-6-luna (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-swe-agent","reasoning_effort":"low","service_mode":null,"score":1.5486725663716814,"score_low":0.7183528101660299,"score_high":2.3789923225773326,"mean_cost_usd_per_task":0.07240623938053097,"median_cost_usd_per_task":0.0697851,"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checked_at":"2026-09-12","archive":"scores-deepswe-1.1-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"gpt-5.6-luna","variantLabel":"mini-swe-agent - gpt-5-6-luna (low)","score":1.5486725663716814,"secondary":{"agentHarness":"mini-swe-agent","reasoningEffort":"low","ciMinus":0.8303197562056515,"ciPlus":0.8303197562056512,"ciMethod":"95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R))","meanCostUsdPerTask":0.07240623938053097,"medianCostPerTaskUsd":0.0697851,"attempted":452,"passed":7,"tasksAttempted":113,"runs":4,"config":"mini_swe_agent_gpt_5_6_luna_low","costBasis":null},"source":"https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:3d9b43ea1f40d6be","board":"deepswe_1_1","model":"deepseek-v4.1-flash","variant":"mini-SWE - DeepSeek V4.1 Flash (max) [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"mini-SWE","reasoning_effort":"max","service_mode":null,"score":74.2,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","checked_at":"2026-09-12","archive":"scores-deepswe-selfreport-2026-09-12.json","raw_record":{"boardId":"deepswe_1_1","model":"deepseek-v4.1-flash","variantLabel":"mini-SWE - DeepSeek V4.1 Flash (max)","score":74.2,"secondary":{"selfReported":true,"reportedBy":"DeepSeek","agentHarness":"mini-SWE","reasoningEffort":"max","contextWindowTokens":1000000,"temperature":1,"topP":0.95},"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","checkedAt":"2026-09-12"}},{"configuration_id":"deepswe_1_1:9454d8de1c52f2fc","board":"deepswe_1_1","model":"mimo-v2.6-pro","variant":"mini-swe-agent - mimo-v2-6-pro [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"mini-swe-agent","reasoning_effort":null,"service_mode":null,"score":71.9,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","checked_at":"2026-09-22","archive":"scores-deepswe-mimo-v26-grok47-selfreport-2026-09-22.json","raw_record":{"boardId":"deepswe_1_1","model":"mimo-v2.6-pro","variantLabel":"mini-swe-agent - mimo-v2-6-pro","score":71.9,"secondary":{"selfReported":true,"reportedBy":"Xiaomi","agentHarness":"mini-swe-agent","agentHarnessAttribution":"inferred-by-comparator-crosscheck","reasoningEffort":null,"effortNote":"MiMo 系列无 reasoning_effort 参数，模型卡未设 effort 档；模型卡推荐采样 temperature=1.0 / top_p=0.95","temperature":1,"topP":0.95,"contextWindowTokens":1000000},"source":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","checkedAt":"2026-09-22","note":"官方模型卡（MIT）第 3 节 Evaluation Results 表 DeepSWE v1.1 行：Pro 71.9 / Flash 67.9 / V2.5 Pro 19.0。同一模型的 DeepSWE 数字有三个互相矛盾的口径，按只记录不择一的规矩全部登记：①模型卡 71.9（本档采用，与 DeepSeek V4.1 Flash 自报同为模型卡口径）；②官方发布页 mimo.mi.com/docs/en-US/news/latest/v2-6 记 RL 曲线 58.4→72.6；③第三方 orcarouter 读小米 Live 面板终点 72.57（面板图注 mini-swe-agent, avg@3）。差异来自 checkpoint 选取、取整与 avg@k，②③未采用。官方榜无 MiMo 行。"}},{"configuration_id":"deepswe_1_1:c2b10fcdefbd0221","board":"deepswe_1_1","model":"mimo-v2.6-flash","variant":"mini-swe-agent - mimo-v2-6-flash [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"mini-swe-agent","reasoning_effort":null,"service_mode":null,"score":67.9,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","checked_at":"2026-09-22","archive":"scores-deepswe-mimo-v26-grok47-selfreport-2026-09-22.json","raw_record":{"boardId":"deepswe_1_1","model":"mimo-v2.6-flash","variantLabel":"mini-swe-agent - mimo-v2-6-flash","score":67.9,"secondary":{"selfReported":true,"reportedBy":"Xiaomi","agentHarness":"mini-swe-agent","agentHarnessAttribution":"inferred-by-comparator-crosscheck","reasoningEffort":null,"effortNote":"MiMo 系列无 reasoning_effort 参数，模型卡未设 effort 档；模型卡推荐采样 temperature=1.0 / top_p=0.95","temperature":1,"topP":0.95,"contextWindowTokens":1000000},"source":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","checkedAt":"2026-09-22","note":"官方模型卡同表 Flash 行 67.9。同型矛盾口径：官方发布页记 RL 曲线 48.8→65.7、第三方读面板终点 65.68，均未采用（理由见 Pro 条）。这是 mimo-v2.6-flash 在本项目七张榜上的第一个分数——此前它 11 个采用点因无分数不进任何能力图。官方榜无 MiMo 行。"}},{"configuration_id":"deepswe_1_1:8fa241178a2088a2","board":"deepswe_1_1","model":"mimo-v2.5-pro","variant":"mini-swe-agent - mimo-v2-5-pro [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"mini-swe-agent","reasoning_effort":null,"service_mode":null,"score":19,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","checked_at":"2026-09-22","archive":"scores-deepswe-mimo-v26-grok47-selfreport-2026-09-22.json","raw_record":{"boardId":"deepswe_1_1","model":"mimo-v2.5-pro","variantLabel":"mini-swe-agent - mimo-v2-5-pro","score":19,"secondary":{"selfReported":true,"reportedBy":"Xiaomi","agentHarness":"mini-swe-agent","agentHarnessAttribution":"inferred-by-comparator-crosscheck","reasoningEffort":null,"effortNote":"MiMo 系列无 reasoning_effort 参数，模型卡未设 effort 档","temperature":1,"topP":0.95},"source":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL","checkedAt":"2026-09-22","note":"同表 MiMo-V2.5 Pro 列 19.0。注意这是厂商在展示新版提升时给出的上一代基准值，方向性偏低的动机存在，读者应按厂商自报折价；与该模型已有的官方榜读数（AA 智力 25.9869、Code Arena 1475、Agent Arena -3.57）量级不矛盾。小米未给 mimo-v2.5（非 Pro）的 DeepSWE 数字，故不补。"}},{"configuration_id":"deepswe_1_1:1d433900eaa0bee0","board":"deepswe_1_1","model":"grok-4.7","variant":"mini-swe-agent - grok-4-7 (high) [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"mini-swe-agent","reasoning_effort":"high","service_mode":null,"score":71,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://x.ai/news/grok-4-7","checked_at":"2026-09-22","archive":"scores-deepswe-mimo-v26-grok47-selfreport-2026-09-22.json","raw_record":{"boardId":"deepswe_1_1","model":"grok-4.7","variantLabel":"mini-swe-agent - grok-4-7 (high)","score":71,"secondary":{"selfReported":true,"reportedBy":"xAI","agentHarness":"mini-swe-agent","agentHarnessAttribution":"inferred-by-comparator-crosscheck","reasoningEffort":"high","effortNote":"发布页基准表整列为 xHigh，DeepSWE 行带星号单独注明 high effort，故本行 effort 为 high 而非 xhigh"},"source":"https://x.ai/news/grok-4-7","checkedAt":"2026-09-22","note":"xAI 发布页（2026-09-21）基准表 DeepSWE v1.1 行 71.0%*，星注 high effort。同表 Grok 4.6 High 65.2 与官方榜 mini-swe-agent - grok-4-6 (high) 65.188 差 0.012，据此判定同 harness 同口径。官方榜无 grok-4.7 行，该模型此前在 DeepSWE 榜缺位。同表 Terminal-Bench 4.0 自报 38.0% 不采用——该榜已有官方行 Grok Build - Grok 4.7 (xhigh) 37.6%，官方读数优先于厂商自报；同表 Grok 4.6 TB4 20.3 与官方榜 Grok Build - Grok 4.6 (high) 20.3 完全一致，进一步佐证两表同源。"}},{"configuration_id":"aa_intelligence_index:4748ef5e83f540ec","board":"aa_intelligence_index","model":"claude-opus-5.5","variant":"Claude Opus 5.5 (max with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":57.6224,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5.5 (max with fallback)","score":57.6224,"secondary":{"configurationRank":1,"costPerTaskUsd":5.982,"scoreRoundedDisplay":58,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-5","shortName":"Claude Opus 5.5 (max with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"max","terminalBench40":0.595959595959596,"scicode":0.668981481481482,"hle":0.613531047265987},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:6e137e293fb02d90","board":"aa_intelligence_index","model":"claude-opus-5.5","variant":"Claude Opus 5.5 (xhigh with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":55.9874,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5.5 (xhigh with fallback)","score":55.9874,"secondary":{"configurationRank":2,"costPerTaskUsd":3.4591,"scoreRoundedDisplay":56,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-5-xhigh","shortName":"Claude Opus 5.5 (xhigh with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"xhigh","terminalBench40":0.595959595959596,"scicode":0.650462962962963,"hle":0.575069508804449},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:40ab71604b0191cd","board":"aa_intelligence_index","model":"claude-sonnet-5.5","variant":"Claude Sonnet 5.5 (max with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":55.978,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5.5 (max with fallback)","score":55.978,"secondary":{"configurationRank":3,"costPerTaskUsd":7.6206,"scoreRoundedDisplay":56,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-5","shortName":"Claude Sonnet 5.5 (max with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"max","terminalBench40":0.636363636363636,"scicode":0.609953703703704,"hle":0.549582947173309},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b65a95fe350f70fe","board":"aa_intelligence_index","model":"claude-opus-5.5","variant":"Claude Opus 5.5 (high with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":53.5832,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5.5 (high with fallback)","score":53.5832,"secondary":{"configurationRank":4,"costPerTaskUsd":1.8225,"scoreRoundedDisplay":54,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-5-high","shortName":"Claude Opus 5.5 (high with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"high","terminalBench40":0.565656565656566,"scicode":0.604166666666667,"hle":0.555607043558851},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:71823637d973a0fc","board":"aa_intelligence_index","model":"claude-fable-5.1","variant":"Claude Fable 5.1 (max with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":53.3549,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_intelligence_index","variantLabel":"Claude Fable 5.1 (max with fallback)","score":53.3549,"secondary":{"configurationRank":5,"costPerTaskUsd":7.6297,"scoreRoundedDisplay":53,"intelligenceIndexIsEstimated":false,"slug":"claude-fable-5-1","shortName":"Claude Fable 5.1 (max with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"max","terminalBench40":0.52020202020202,"scicode":0.630787037037037,"hle":0.591288229842447},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:3f66aafbb8812925","board":"aa_intelligence_index","model":"claude-fable-5.1","variant":"Claude Fable 5.1 (xhigh with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":53.2033,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_intelligence_index","variantLabel":"Claude Fable 5.1 (xhigh with fallback)","score":53.2033,"secondary":{"configurationRank":6,"costPerTaskUsd":5.9783,"scoreRoundedDisplay":53,"intelligenceIndexIsEstimated":false,"slug":"claude-fable-5-1-xhigh","shortName":"Claude Fable 5.1 (xhigh with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"xhigh","terminalBench40":0.55050505050505,"scicode":0.608796296296296,"hle":0.587117701575533},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:9b1da82a91856d8d","board":"aa_intelligence_index","model":"gpt-6-astra","variant":"GPT-6 Astra (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":52.6737,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Astra (max)","score":52.6737,"secondary":{"configurationRank":7,"costPerTaskUsd":3.2575,"scoreRoundedDisplay":53,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-astra","shortName":"GPT-6 Astra (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.590909090909091,"scicode":0.564814814814815,"hle":0.546802594995366},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:7070b585da6e3a29","board":"aa_intelligence_index","model":"gpt-6-astra","variant":"GPT-6 Astra (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":52.3863,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Astra (xhigh)","score":52.3863,"secondary":{"configurationRank":9,"costPerTaskUsd":2.3088,"scoreRoundedDisplay":52,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-astra-xhigh","shortName":"GPT-6 Astra (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.595959595959596,"scicode":0.556712962962963,"hle":0.545875810936052},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:87e4e17f2fac9de8","board":"aa_intelligence_index","model":"claude-sonnet-5.5","variant":"Claude Sonnet 5.5 (xhigh with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":51.8523,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5.5 (xhigh with fallback)","score":51.8523,"secondary":{"configurationRank":10,"costPerTaskUsd":2.7426,"scoreRoundedDisplay":52,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-5-xhigh","shortName":"Claude Sonnet 5.5 (xhigh with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"xhigh","terminalBench40":0.570707070707071,"scicode":0.572916666666667,"hle":0.5},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b6ad871a45a4a5df","board":"aa_intelligence_index","model":"gpt-6.1-sol","variant":"GPT-6.1 Sol (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":51.8333,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6.1 Sol (max)","score":51.8333,"secondary":{"configurationRank":11,"costPerTaskUsd":0.7242,"scoreRoundedDisplay":52,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-1-sol","shortName":"GPT-6.1 Sol (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.560606060606061,"scicode":0.541666666666667,"hle":0.529193697868397},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:6b86f0e5c649cf41","board":"aa_intelligence_index","model":"claude-opus-5.5","variant":"Claude Opus 5.5 (medium with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":51.2435,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5.5 (medium with fallback)","score":51.2435,"secondary":{"configurationRank":12,"costPerTaskUsd":1.336,"scoreRoundedDisplay":51,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-5-medium","shortName":"Claude Opus 5.5 (medium with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"medium","terminalBench40":0.525252525252525,"scicode":0.592592592592593,"hle":0.547265987025023},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:74265ef58fb6e45b","board":"aa_intelligence_index","model":"claude-fable-5.1","variant":"Claude Fable 5.1 (high with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":51.1516,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_intelligence_index","variantLabel":"Claude Fable 5.1 (high with fallback)","score":51.1516,"secondary":{"configurationRank":13,"costPerTaskUsd":3.9125,"scoreRoundedDisplay":51,"intelligenceIndexIsEstimated":false,"slug":"claude-fable-5-1-high","shortName":"Claude Fable 5.1 (high with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"high","terminalBench40":0.52020202020202,"scicode":0.586805555555556,"hle":0.559314179796108},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:4a8f2c60cc909b11","board":"aa_intelligence_index","model":"gpt-6.1-sol","variant":"GPT-6.1 Sol (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":51.0378,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6.1 Sol (xhigh)","score":51.0378,"secondary":{"configurationRank":14,"costPerTaskUsd":0.3929,"scoreRoundedDisplay":51,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-1-sol-xhigh","shortName":"GPT-6.1 Sol (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.54040404040404,"scicode":0.556712962962963,"hle":0.526413345690454},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:7c09034755a992fc","board":"aa_intelligence_index","model":"gpt-6-astra","variant":"GPT-6 Astra (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":50.9191,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Astra (high)","score":50.9191,"secondary":{"configurationRank":15,"costPerTaskUsd":1.7253,"scoreRoundedDisplay":51,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-astra-high","shortName":"GPT-6 Astra (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.54040404040404,"scicode":0.554398148148148,"hle":0.530583873957368},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:68425237e9876779","board":"aa_intelligence_index","model":"claude-opus-5","variant":"Claude Opus 5 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":50.7771,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5 (max)","score":50.7771,"secondary":{"configurationRank":16,"costPerTaskUsd":5.8584,"scoreRoundedDisplay":51,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5","shortName":"Claude Opus 5 (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"max","terminalBench40":0.48989898989899,"scicode":0.563657407407407,"hle":0.548656163113994},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:aea0241d837a4380","board":"aa_intelligence_index","model":"gpt-6.1-sol","variant":"GPT-6.1 Sol (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":50.2378,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6.1 Sol (high)","score":50.2378,"secondary":{"configurationRank":17,"costPerTaskUsd":0.3191,"scoreRoundedDisplay":50,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-1-sol-high","shortName":"GPT-6.1 Sol (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.515151515151515,"scicode":0.55787037037037,"hle":0.513901760889713},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:3a154cb4e9c99189","board":"aa_intelligence_index","model":"claude-opus-5","variant":"Claude Opus 5 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":49.6774,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5 (xhigh)","score":49.6774,"secondary":{"configurationRank":18,"costPerTaskUsd":4.8778,"scoreRoundedDisplay":50,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-xhigh","shortName":"Claude Opus 5 (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"xhigh","terminalBench40":0.464646464646465,"scicode":0.556712962962963,"hle":0.544022242817424},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:9b5f20f46c907550","board":"aa_intelligence_index","model":"claude-fable-5","variant":"Claude Fable 5 (with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":49.6258,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5","boardId":"aa_intelligence_index","variantLabel":"Claude Fable 5 (with fallback)","score":49.6258,"secondary":{"configurationRank":19,"costPerTaskUsd":8.746,"scoreRoundedDisplay":50,"intelligenceIndexIsEstimated":false,"slug":"claude-fable-5","shortName":"Claude Fable 5 (with fallback)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":null,"terminalBench40":0.424242424242424,"scicode":0.609953703703704,"hle":0.554680259499537},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:3cf95b4f1be42e0f","board":"aa_intelligence_index","model":"gpt-6-astra","variant":"GPT-6 Astra (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":49.5704,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Astra (medium)","score":49.5704,"secondary":{"configurationRank":20,"costPerTaskUsd":1.5406,"scoreRoundedDisplay":50,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-astra-medium","shortName":"GPT-6 Astra (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.494949494949495,"scicode":0.541666666666667,"hle":0.527340129749768},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:9a927f4404e32355","board":"aa_intelligence_index","model":"claude-fable-5.1","variant":"Claude Fable 5.1 (medium with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":48.9206,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_intelligence_index","variantLabel":"Claude Fable 5.1 (medium with fallback)","score":48.9206,"secondary":{"configurationRank":21,"costPerTaskUsd":2.9826,"scoreRoundedDisplay":49,"intelligenceIndexIsEstimated":false,"slug":"claude-fable-5-1-medium","shortName":"Claude Fable 5.1 (medium with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"medium","terminalBench40":0.44949494949495,"scicode":0.563657407407407,"hle":0.537998146431881},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:295f5f7bdfa8d248","board":"aa_intelligence_index","model":"claude-opus-5","variant":"Claude Opus 5 (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":48.1219,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5 (high)","score":48.1219,"secondary":{"configurationRank":22,"costPerTaskUsd":3.6133,"scoreRoundedDisplay":48,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-high","shortName":"Claude Opus 5 (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"high","terminalBench40":0.45959595959596,"scicode":0.554398148148148,"hle":0.528266913809083},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:a8c28362dff3aa31","board":"aa_intelligence_index","model":"muse-spark-1.3","variant":"Muse Spark 1.3 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":48.0923,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.3","boardId":"aa_intelligence_index","variantLabel":"Muse Spark 1.3 (max)","score":48.0923,"secondary":{"configurationRank":23,"costPerTaskUsd":1.6049,"scoreRoundedDisplay":48,"intelligenceIndexIsEstimated":false,"slug":"muse-spark-1-3","shortName":"Muse Spark 1.3 (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"Meta","effort":"max","terminalBench40":0.333333333333333,"scicode":0.587962962962963,"hle":0.487025023169602},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:e4a9f446e946fc59","board":"aa_intelligence_index","model":"gpt-6.1-sol","variant":"GPT-6.1 Sol (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":47.7833,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6.1 Sol (medium)","score":47.7833,"secondary":{"configurationRank":24,"costPerTaskUsd":0.2137,"scoreRoundedDisplay":48,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-1-sol-medium","shortName":"GPT-6.1 Sol (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.47979797979798,"scicode":0.532407407407407,"hle":0.498609823911029},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:ca545106241767d7","board":"aa_intelligence_index","model":"gpt-6-sol","variant":"GPT-6 Sol (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":47.5276,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Sol (max)","score":47.5276,"secondary":{"configurationRank":25,"costPerTaskUsd":1.0494,"scoreRoundedDisplay":48,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-sol","shortName":"GPT-6 Sol (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.439393939393939,"scicode":0.576388888888889,"hle":0.479147358665431},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c95dc5141f32b497","board":"aa_intelligence_index","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":46.9727,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Sol (max)","score":46.9727,"secondary":{"configurationRank":26,"costPerTaskUsd":1.9885,"scoreRoundedDisplay":47,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-sol","shortName":"GPT-5.6 Sol (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.398989898989899,"scicode":0.570601851851852,"hle":0.494902687673772},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2ceeec877967a38a","board":"aa_intelligence_index","model":"claude-fable-5.1","variant":"Claude Fable 5.1 (low with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":46.8163,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_intelligence_index","variantLabel":"Claude Fable 5.1 (low with fallback)","score":46.8163,"secondary":{"configurationRank":27,"costPerTaskUsd":2.371,"scoreRoundedDisplay":47,"intelligenceIndexIsEstimated":false,"slug":"claude-fable-5-1-low","shortName":"Claude Fable 5.1 (low with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"low","terminalBench40":0.404040404040404,"scicode":0.56712962962963,"hle":0.48887859128823},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:07dd9777aab0951b","board":"aa_intelligence_index","model":"claude-sonnet-5.5","variant":"Claude Sonnet 5.5 (high with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":46.7357,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5.5 (high with fallback)","score":46.7357,"secondary":{"configurationRank":28,"costPerTaskUsd":1.0798,"scoreRoundedDisplay":47,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-5-high","shortName":"Claude Sonnet 5.5 (high with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"high","terminalBench40":0.439393939393939,"scicode":0.537037037037037,"hle":0.458294717330862},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:d11843ec849da6d9","board":"aa_intelligence_index","model":"grok-4.7","variant":"Grok 4.7 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":46.4466,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.7","boardId":"aa_intelligence_index","variantLabel":"Grok 4.7 (xhigh)","score":46.4466,"secondary":{"configurationRank":29,"costPerTaskUsd":3.7383,"scoreRoundedDisplay":46,"intelligenceIndexIsEstimated":false,"slug":"grok-4-7","shortName":"Grok 4.7 (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"SpaceXAI","effort":"xhigh","terminalBench40":0.257575757575758,"scicode":0.574074074074074,"hle":0.431417979610751},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:0c727f941b7eae61","board":"aa_intelligence_index","model":"grok-4.7","variant":"Grok 4.7 (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":46.3322,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.7","boardId":"aa_intelligence_index","variantLabel":"Grok 4.7 (high)","score":46.3322,"secondary":{"configurationRank":30,"costPerTaskUsd":2.7261,"scoreRoundedDisplay":46,"intelligenceIndexIsEstimated":false,"slug":"grok-4-7-high","shortName":"Grok 4.7 (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"SpaceXAI","effort":"high","terminalBench40":0.247474747474747,"scicode":0.577546296296296,"hle":0.423076923076923},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:83bd852b96aedbce","board":"aa_intelligence_index","model":"mimo-v2.6-pro","variant":"MiMo-V2.6-Pro","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":46.3242,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.6-pro","boardId":"aa_intelligence_index","variantLabel":"MiMo-V2.6-Pro","score":46.3242,"secondary":{"configurationRank":31,"costPerTaskUsd":0.1332,"scoreRoundedDisplay":46,"intelligenceIndexIsEstimated":false,"slug":"mimo-v2-6-pro","shortName":"MiMo-V2.6-Pro","isReasoning":true,"deprecated":false,"modelCreatorName":"Xiaomi","effort":null,"terminalBench40":0.348484848484849,"scicode":0.608796296296296,"hle":0.493512511584801},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:6858c62f40a2b8a3","board":"aa_intelligence_index","model":"gpt-6-astra","variant":"GPT-6 Astra (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":45.7819,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Astra (low)","score":45.7819,"secondary":{"configurationRank":32,"costPerTaskUsd":0.8175,"scoreRoundedDisplay":46,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-astra-low","shortName":"GPT-6 Astra (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0.419191919191919,"scicode":0.540509259259259,"hle":0.492122335495829},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c9c2272b5b7ddbdf","board":"aa_intelligence_index","model":"qwen3.8-max-0902","variant":"Qwen3.8 Max (0902)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":45.4152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-max-0902","boardId":"aa_intelligence_index","variantLabel":"Qwen3.8 Max (0902)","score":45.4152,"secondary":{"configurationRank":33,"costPerTaskUsd":5.4085,"scoreRoundedDisplay":45,"intelligenceIndexIsEstimated":false,"slug":"qwen3-8-max","shortName":"Qwen3.8 Max (0902)","isReasoning":true,"deprecated":false,"modelCreatorName":"Alibaba","effort":"max","terminalBench40":0.388888888888889,"scicode":0.520833333333333,"hle":0.430954587581094},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:eaee770ce12a30df","board":"aa_intelligence_index","model":"muse-spark-1.3","variant":"Muse Spark 1.3 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":45.0733,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.3","boardId":"aa_intelligence_index","variantLabel":"Muse Spark 1.3 (xhigh)","score":45.0733,"secondary":{"configurationRank":34,"costPerTaskUsd":1.3678,"scoreRoundedDisplay":45,"intelligenceIndexIsEstimated":false,"slug":"muse-spark-1-3-xhigh","shortName":"Muse Spark 1.3 (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"Meta","effort":"xhigh","terminalBench40":0.166666666666667,"scicode":0.597222222222222,"hle":0.47451343836886},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:bacfec5f5f91d13c","board":"aa_intelligence_index","model":"claude-opus-5","variant":"Claude Opus 5 (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":44.8253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5 (medium)","score":44.8253,"secondary":{"configurationRank":35,"costPerTaskUsd":2.1895,"scoreRoundedDisplay":45,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-medium","shortName":"Claude Opus 5 (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"medium","terminalBench40":0.343434343434343,"scicode":0.515046296296296,"hle":0.512974976830398},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:21922ac966701739","board":"aa_intelligence_index","model":"glm-5.3","variant":"GLM-5.3 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":44.7774,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3","boardId":"aa_intelligence_index","variantLabel":"GLM-5.3 (max)","score":44.7774,"secondary":{"configurationRank":36,"costPerTaskUsd":2.0056,"scoreRoundedDisplay":45,"intelligenceIndexIsEstimated":false,"slug":"glm-5-3","shortName":"GLM-5.3 (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"Z AI","effort":"max","terminalBench40":0.419191919191919,"scicode":0.590277777777778,"hle":0.422613531047266},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b56103ded435971f","board":"aa_intelligence_index","model":"grok-4.6","variant":"Grok 4.6 (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":44.3113,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_intelligence_index","variantLabel":"Grok 4.6 (high)","score":44.3113,"secondary":{"configurationRank":37,"costPerTaskUsd":1.8589,"scoreRoundedDisplay":44,"intelligenceIndexIsEstimated":false,"slug":"grok-4-6","shortName":"Grok 4.6 (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"SpaceXAI","effort":"high","terminalBench40":0.212121212121212,"scicode":0.564814814814815,"hle":0.429101019462465},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:13ecd3c1725d233e","board":"aa_intelligence_index","model":"grok-4.6","variant":"Grok 4.6 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":44.1998,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_intelligence_index","variantLabel":"Grok 4.6 (xhigh)","score":44.1998,"secondary":{"configurationRank":38,"costPerTaskUsd":2.3237,"scoreRoundedDisplay":44,"intelligenceIndexIsEstimated":false,"slug":"grok-4-6-xhigh","shortName":"Grok 4.6 (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"SpaceXAI","effort":"xhigh","terminalBench40":0.171717171717172,"scicode":0.530092592592593,"hle":0.440685820203893},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:ec3291a5ccf1d360","board":"aa_intelligence_index","model":"gpt-6-sol","variant":"GPT-6 Sol (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":44.101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Sol (xhigh)","score":44.101,"secondary":{"configurationRank":39,"costPerTaskUsd":0.5236,"scoreRoundedDisplay":44,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-sol-xhigh","shortName":"GPT-6 Sol (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.303030303030303,"scicode":0.550925925925926,"hle":0.462928637627433},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:467034ea7730fe43","board":"aa_intelligence_index","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":44.0089,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Sol (xhigh)","score":44.0089,"secondary":{"configurationRank":40,"costPerTaskUsd":1.1844,"scoreRoundedDisplay":44,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-sol-xhigh","shortName":"GPT-5.6 Sol (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.247474747474747,"scicode":0.570601851851852,"hle":0.473123262279889},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:fb0405a168316f85","board":"aa_intelligence_index","model":"step-5-preview","variant":"Step 5 Preview","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":43.7343,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"step-5-preview","boardId":"aa_intelligence_index","variantLabel":"Step 5 Preview","score":43.7343,"secondary":{"configurationRank":41,"costPerTaskUsd":0.7175,"scoreRoundedDisplay":44,"intelligenceIndexIsEstimated":false,"slug":"step-5","shortName":"Step 5 Preview","isReasoning":true,"deprecated":false,"modelCreatorName":"StepFun","effort":null,"terminalBench40":0.333333333333333,"scicode":0.58912037037037,"hle":0.464782205746061},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:13115cf53ac21275","board":"aa_intelligence_index","model":"kimi-k3","variant":"Kimi K3 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":43.5938,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k3","boardId":"aa_intelligence_index","variantLabel":"Kimi K3 (max)","score":43.5938,"secondary":{"configurationRank":42,"costPerTaskUsd":2.0001,"scoreRoundedDisplay":44,"intelligenceIndexIsEstimated":false,"slug":"kimi-k3","shortName":"Kimi K3 (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"Kimi","effort":"max","terminalBench40":0.126262626262626,"scicode":0.594907407407407,"hle":0.468952734012975},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2df81011d518b272","board":"aa_intelligence_index","model":"grok-4.6","variant":"Grok 4.6 (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":42.8363,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_intelligence_index","variantLabel":"Grok 4.6 (medium)","score":42.8363,"secondary":{"configurationRank":43,"costPerTaskUsd":1.4963,"scoreRoundedDisplay":43,"intelligenceIndexIsEstimated":false,"slug":"grok-4-6-medium","shortName":"Grok 4.6 (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"SpaceXAI","effort":"medium","terminalBench40":0.131313131313131,"scicode":0.559027777777778,"hle":0.421223354958295},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b88d90b9ef2bcc13","board":"aa_intelligence_index","model":"gpt-6-sol","variant":"GPT-6 Sol (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":42.8216,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Sol (high)","score":42.8216,"secondary":{"configurationRank":44,"costPerTaskUsd":0.3753,"scoreRoundedDisplay":43,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-sol-high","shortName":"GPT-6 Sol (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.262626262626263,"scicode":0.548611111111111,"hle":0.44114921223355},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:5b0f75c30f0ab19b","board":"aa_intelligence_index","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":42.3461,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Sol (high)","score":42.3461,"secondary":{"configurationRank":45,"costPerTaskUsd":0.808,"scoreRoundedDisplay":42,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-sol-high","shortName":"GPT-5.6 Sol (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.207070707070707,"scicode":0.577546296296296,"hle":0.46014828544949},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:402fce0f28908149","board":"aa_intelligence_index","model":"claude-opus-5.5","variant":"Claude Opus 5.5 (low with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":42.3078,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5.5 (low with fallback)","score":42.3078,"secondary":{"configurationRank":46,"costPerTaskUsd":0.5512,"scoreRoundedDisplay":42,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-5-low","shortName":"Claude Opus 5.5 (low with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"low","terminalBench40":0.313131313131313,"scicode":0.585648148148148,"hle":0.483317886932345},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:f49734a44b6bfe83","board":"aa_intelligence_index","model":"gpt-6.1-sol","variant":"GPT-6.1 Sol (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":42.0836,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6.1 Sol (low)","score":42.0836,"secondary":{"configurationRank":47,"costPerTaskUsd":0.1308,"scoreRoundedDisplay":42,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-1-sol-low","shortName":"GPT-6.1 Sol (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0.308080808080808,"scicode":0.532407407407407,"hle":0.473586654309546},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:caaac19d5e8e542f","board":"aa_intelligence_index","model":"gpt-5.6-terra","variant":"GPT-5.6 Terra (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":42.0829,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Terra (max)","score":42.0829,"secondary":{"configurationRank":48,"costPerTaskUsd":1.3987,"scoreRoundedDisplay":42,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-terra","shortName":"GPT-5.6 Terra (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.353535353535354,"scicode":0.549768518518518,"hle":0.429101019462465},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:066a860a43b7204e","board":"aa_intelligence_index","model":"glm-5.3-flash","variant":"GLM-5.3-Flash","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":41.8075,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3-flash","boardId":"aa_intelligence_index","variantLabel":"GLM-5.3-Flash","score":41.8075,"secondary":{"configurationRank":49,"costPerTaskUsd":0.2533,"scoreRoundedDisplay":42,"intelligenceIndexIsEstimated":false,"slug":"glm-5-3-flash","shortName":"GLM-5.3-Flash","isReasoning":true,"deprecated":false,"modelCreatorName":"Z AI","effort":null,"terminalBench40":0.328282828282828,"scicode":0.516203703703704,"hle":0.398517145505097},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:8ba60e513dcc3861","board":"aa_intelligence_index","model":"claude-opus-4.8","variant":"Claude Opus 4.8 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":41.7899,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-4.8","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 4.8 (max)","score":41.7899,"secondary":{"configurationRank":50,"costPerTaskUsd":4.081,"scoreRoundedDisplay":42,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-4-8","shortName":"Claude Opus 4.8 (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"max","terminalBench40":0.217171717171717,"scicode":0.543981481481482,"hle":0.486561631139944},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:a795ed51d673418d","board":"aa_intelligence_index","model":"gemini-3.8-flash","variant":"Gemini 3.8 Flash (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":40.9262,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.8 Flash (high)","score":40.9262,"secondary":{"configurationRank":51,"costPerTaskUsd":1.2428,"scoreRoundedDisplay":41,"intelligenceIndexIsEstimated":false,"slug":"gemini-3-8-flash","shortName":"Gemini 3.8 Flash (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"Google","effort":"high","terminalBench40":0.196969696969697,"scicode":0.565972222222222,"hle":0.478220574606117},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:328c4eadb5cbefe8","board":"aa_intelligence_index","model":"claude-sonnet-5.5","variant":"Claude Sonnet 5.5 (medium with fallback)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":40.7387,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5.5 (medium with fallback)","score":40.7387,"secondary":{"configurationRank":52,"costPerTaskUsd":0.5862,"scoreRoundedDisplay":41,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-5-medium","shortName":"Claude Sonnet 5.5 (medium with fallback)","isReasoning":true,"deprecated":false,"modelCreatorName":"Anthropic","effort":"medium","terminalBench40":0.297979797979798,"scicode":0.528935185185185,"hle":0.39805375347544},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:8919405f90e7565f","board":"aa_intelligence_index","model":"qwen3.8-max","variant":"Qwen3.8 Max","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":40.1529,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-max","boardId":"aa_intelligence_index","variantLabel":"Qwen3.8 Max","score":40.1529,"secondary":{"configurationRank":54,"costPerTaskUsd":2.67,"scoreRoundedDisplay":40,"intelligenceIndexIsEstimated":false,"slug":"qwen3-8-max-0803","shortName":"Qwen3.8 Max","isReasoning":true,"deprecated":true,"modelCreatorName":"Alibaba","effort":null,"terminalBench40":0.186868686868687,"scicode":0.532407407407407,"hle":0.430491195551437},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:5dea08c56cadf0a5","board":"aa_intelligence_index","model":"gpt-6-sol","variant":"GPT-6 Sol (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":39.7821,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Sol (medium)","score":39.7821,"secondary":{"configurationRank":57,"costPerTaskUsd":0.2472,"scoreRoundedDisplay":40,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-sol-medium","shortName":"GPT-6 Sol (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.186868686868687,"scicode":0.538194444444444,"hle":0.409638554216867},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:8f226080bd0bde9a","board":"aa_intelligence_index","model":"gemini-3.8-flash","variant":"Gemini 3.8 Flash (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":39.774,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.8 Flash (medium)","score":39.774,"secondary":{"configurationRank":58,"costPerTaskUsd":0.931,"scoreRoundedDisplay":40,"intelligenceIndexIsEstimated":false,"slug":"gemini-3-8-flash-medium","shortName":"Gemini 3.8 Flash (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"Google","effort":"medium","terminalBench40":0.196969696969697,"scicode":0.550925925925926,"hle":0.421223354958295},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:59269287bc1e95d8","board":"aa_intelligence_index","model":"gemini-3.7-flash","variant":"Gemini 3.7 Flash (medium) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":39.6178,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.7-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.7 Flash (medium)","score":39.6178,"secondary":{"configurationRank":59,"costPerTaskUsd":null,"scoreRoundedDisplay":40,"intelligenceIndexIsEstimated":true,"slug":"gemini-3-7-flash-medium","shortName":"Gemini 3.7 Flash (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"Google","effort":"medium","terminalBench40":null,"scicode":0.59837962962963,"hle":0.389712696941613},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:9666106dbc0fffc1","board":"aa_intelligence_index","model":"muse-spark-1.2","variant":"Muse Spark 1.2 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":39.5759,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.2","boardId":"aa_intelligence_index","variantLabel":"Muse Spark 1.2 (xhigh)","score":39.5759,"secondary":{"configurationRank":60,"costPerTaskUsd":0.9747,"scoreRoundedDisplay":40,"intelligenceIndexIsEstimated":false,"slug":"muse-spark-1-2","shortName":"Muse Spark 1.2 (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"Meta","effort":"xhigh","terminalBench40":0.0707070707070707,"scicode":0.574074074074074,"hle":0.454587581093605},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:8076c0cba16368e2","board":"aa_intelligence_index","model":"deepseek-v4.1-flash","variant":"DeepSeek V4.1 Flash (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":39.4562,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4.1-flash","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4.1 Flash (max)","score":39.4562,"secondary":{"configurationRank":61,"costPerTaskUsd":0.2652,"scoreRoundedDisplay":39,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-1-flash","shortName":"DeepSeek V4.1 Flash (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"DeepSeek","effort":"max","terminalBench40":0.267676767676768,"scicode":0.518518518518518,"hle":0.392493049119555},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:e42133cf9100a842","board":"aa_intelligence_index","model":"claude-opus-5","variant":"Claude Opus 5 (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":39.3543,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_intelligence_index","variantLabel":"Claude Opus 5 (low)","score":39.3543,"secondary":{"configurationRank":62,"costPerTaskUsd":1.0983,"scoreRoundedDisplay":39,"intelligenceIndexIsEstimated":false,"slug":"claude-opus-5-low","shortName":"Claude Opus 5 (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"low","terminalBench40":0.262626262626263,"scicode":0.491898148148148,"hle":0.434198331788693},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:64efdbd82ca66394","board":"aa_intelligence_index","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":39.2364,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Sol (medium)","score":39.2364,"secondary":{"configurationRank":63,"costPerTaskUsd":0.505,"scoreRoundedDisplay":39,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-sol-medium","shortName":"GPT-5.6 Sol (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.146464646464646,"scicode":0.574074074074074,"hle":0.422150139017609},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:aeabf6017dd74bc8","board":"aa_intelligence_index","model":"gemini-3.7-flash","variant":"Gemini 3.7 Flash (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":39.0595,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.7-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.7 Flash (high)","score":39.0595,"secondary":{"configurationRank":64,"costPerTaskUsd":0.9253,"scoreRoundedDisplay":39,"intelligenceIndexIsEstimated":false,"slug":"gemini-3-7-flash","shortName":"Gemini 3.7 Flash (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"Google","effort":"high","terminalBench40":0.136363636363636,"scicode":0.571759259259259,"hle":0.478683966635774},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:829ff4209a0293bc","board":"aa_intelligence_index","model":"grok-4.5","variant":"Grok 4.5 (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":38.8121,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.5","boardId":"aa_intelligence_index","variantLabel":"Grok 4.5 (high)","score":38.8121,"secondary":{"configurationRank":66,"costPerTaskUsd":1.0365,"scoreRoundedDisplay":39,"intelligenceIndexIsEstimated":false,"slug":"grok-4-5","shortName":"Grok 4.5 (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"SpaceXAI","effort":"high","terminalBench40":0.106060606060606,"scicode":0.549768518518518,"hle":0.42678405931418},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c4bfcb80a46cebef","board":"aa_intelligence_index","model":"gpt-5.5","variant":"GPT-5.5 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":38.3556,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_intelligence_index","variantLabel":"GPT-5.5 (xhigh)","score":38.3556,"secondary":{"configurationRank":67,"costPerTaskUsd":2.634,"scoreRoundedDisplay":38,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-5","shortName":"GPT-5.5 (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.146464646464646,"scicode":0.55787037037037,"hle":0.457831325301205},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:00e5e17098f2a9cc","board":"aa_intelligence_index","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":38.1639,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5 (max)","score":38.1639,"secondary":{"configurationRank":68,"costPerTaskUsd":5.0912,"scoreRoundedDisplay":38,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5","shortName":"Claude Sonnet 5 (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"max","terminalBench40":0.141414141414141,"scicode":0.542824074074074,"hle":0.412882298424467},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:660875fc06b206b1","board":"aa_intelligence_index","model":"gpt-5.6-terra","variant":"GPT-5.6 Terra (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":37.9513,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Terra (xhigh)","score":37.9513,"secondary":{"configurationRank":69,"costPerTaskUsd":0.6318,"scoreRoundedDisplay":38,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-terra-xhigh","shortName":"GPT-5.6 Terra (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.101010101010101,"scicode":0.523148148148148,"hle":0.418906394810009},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2446e4ad7216a14c","board":"aa_intelligence_index","model":"mimo-v2.6-flash","variant":"MiMo-V2.6-Flash","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":37.8844,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.6-flash","boardId":"aa_intelligence_index","variantLabel":"MiMo-V2.6-Flash","score":37.8844,"secondary":{"configurationRank":70,"costPerTaskUsd":0.0622,"scoreRoundedDisplay":38,"intelligenceIndexIsEstimated":false,"slug":"mimo-v2-6-flash","shortName":"MiMo-V2.6-Flash","isReasoning":true,"deprecated":false,"modelCreatorName":"Xiaomi","effort":null,"terminalBench40":0.227272727272727,"scicode":0.512731481481482,"hle":0.350787766450417},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:6477a68e40370653","board":"aa_intelligence_index","model":"gpt-5.6-luna","variant":"GPT-5.6 Luna (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":37.3244,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Luna (max)","score":37.3244,"secondary":{"configurationRank":71,"costPerTaskUsd":0.1783,"scoreRoundedDisplay":37,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-luna","shortName":"GPT-5.6 Luna (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.116161616161616,"scicode":0.53587962962963,"hle":0.394810009267841},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:e41e4b6709da24d4","board":"aa_intelligence_index","model":"gpt-6-luna","variant":"GPT-6 Luna (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":37.256,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Luna (max)","score":37.256,"secondary":{"configurationRank":72,"costPerTaskUsd":0.0679,"scoreRoundedDisplay":37,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-luna","shortName":"GPT-6 Luna (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"max","terminalBench40":0.126262626262626,"scicode":0.546296296296296,"hle":0.385078776645042},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:6fd00cd9ccb51b99","board":"aa_intelligence_index","model":"gpt-5.5","variant":"GPT-5.5 (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":36.9794,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_intelligence_index","variantLabel":"GPT-5.5 (high)","score":36.9794,"secondary":{"configurationRank":73,"costPerTaskUsd":1.5413,"scoreRoundedDisplay":37,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-5-high","shortName":"GPT-5.5 (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.0909090909090909,"scicode":0.561342592592593,"hle":0.450417052826691},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:cc014f66e9d183e0","board":"aa_intelligence_index","model":"gemini-3.7-flash","variant":"Gemini 3.7 Flash (low) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":36.9459,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.7-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.7 Flash (low)","score":36.9459,"secondary":{"configurationRank":74,"costPerTaskUsd":null,"scoreRoundedDisplay":37,"intelligenceIndexIsEstimated":true,"slug":"gemini-3-7-flash-low","shortName":"Gemini 3.7 Flash (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"Google","effort":"low","terminalBench40":null,"scicode":0.556712962962963,"hle":0.351251158480074},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:e551b2b658c831ff","board":"aa_intelligence_index","model":"deepseek-v4-pro","variant":"DeepSeek V4 Pro 0813 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":35.9968,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Pro 0813 (max)","score":35.9968,"secondary":{"configurationRank":75,"costPerTaskUsd":0.674,"scoreRoundedDisplay":36,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-pro","shortName":"DeepSeek V4 Pro 0813 (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"DeepSeek","effort":"max","terminalBench40":0.141414141414141,"scicode":0.510416666666667,"hle":0.410101946246525},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2e9f902e3fc120db","board":"aa_intelligence_index","model":"grok-4.6","variant":"Grok 4.6 (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":35.1247,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_intelligence_index","variantLabel":"Grok 4.6 (low)","score":35.1247,"secondary":{"configurationRank":76,"costPerTaskUsd":0.4753,"scoreRoundedDisplay":35,"intelligenceIndexIsEstimated":false,"slug":"grok-4-6-low","shortName":"Grok 4.6 (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"SpaceXAI","effort":"low","terminalBench40":0.0303030303030303,"scicode":0.494212962962963,"hle":0.276181649675626},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2ab3fed6fb264856","board":"aa_intelligence_index","model":"deepseek-v4-flash-vision-exp","variant":"DeepSeek V4 Flash Vision (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":34.8391,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash-vision-exp","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Flash Vision (max)","score":34.8391,"secondary":{"configurationRank":77,"costPerTaskUsd":0.3144,"scoreRoundedDisplay":35,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-flash-vision","shortName":"DeepSeek V4 Flash Vision (max)","isReasoning":true,"deprecated":false,"modelCreatorName":"DeepSeek","effort":"max","terminalBench40":0.121212121212121,"scicode":0.496527777777778,"hle":0.344763670064875},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:cbccf58a680025ea","board":"aa_intelligence_index","model":"gpt-5.6-luna","variant":"GPT-5.6 Luna (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":34.5553,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Luna (xhigh)","score":34.5553,"secondary":{"configurationRank":78,"costPerTaskUsd":0.0853,"scoreRoundedDisplay":35,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-luna-xhigh","shortName":"GPT-5.6 Luna (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.0353535353535354,"scicode":0.50462962962963,"hle":0.369786839666358},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:26fc619ae9c44efb","board":"aa_intelligence_index","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":34.3842,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5 (xhigh)","score":34.3842,"secondary":{"configurationRank":79,"costPerTaskUsd":2.872,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-xhigh","shortName":"Claude Sonnet 5 (xhigh)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"xhigh","terminalBench40":0.0707070707070707,"scicode":0.540509259259259,"hle":0.39017608897127},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b6a092361e695dd6","board":"aa_intelligence_index","model":"deepseek-v4-flash","variant":"DeepSeek V4 Flash 0731 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":34.3305,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Flash 0731 (max)","score":34.3305,"secondary":{"configurationRank":80,"costPerTaskUsd":0.2196,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-flash","shortName":"DeepSeek V4 Flash 0731 (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"max","terminalBench40":0.121212121212121,"scicode":0.503472222222222,"hle":0.385542168674699},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:237ff152b6aa847d","board":"aa_intelligence_index","model":"glm-5.3","variant":"GLM-5.3 (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":34.299,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3","boardId":"aa_intelligence_index","variantLabel":"GLM-5.3 (low)","score":34.299,"secondary":{"configurationRank":81,"costPerTaskUsd":0.8522,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"glm-5-3-low","shortName":"GLM-5.3 (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"Z AI","effort":"low","terminalBench40":0.348484848484849,"scicode":0.420138888888889,"hle":0.365616311399444},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:7d5413682aa8f8af","board":"aa_intelligence_index","model":"gpt-5.6-terra","variant":"GPT-5.6 Terra (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":34.2377,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Terra (high)","score":34.2377,"secondary":{"configurationRank":82,"costPerTaskUsd":0.3379,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-terra-high","shortName":"GPT-5.6 Terra (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.0151515151515152,"scicode":0.524305555555556,"hle":0.385078776645042},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c190b81e19d238cc","board":"aa_intelligence_index","model":"gemini-3.6-flash","variant":"Gemini 3.6 Flash (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":33.9786,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.6-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.6 Flash (high)","score":33.9786,"secondary":{"configurationRank":83,"costPerTaskUsd":0.9288,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"gemini-3-6-flash","shortName":"Gemini 3.6 Flash (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"Google","effort":"high","terminalBench40":0.0707070707070707,"scicode":0.533564814814815,"hle":0.408248378127896},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:db3deb60f0a6f0ac","board":"aa_intelligence_index","model":"gpt-6-sol","variant":"GPT-6 Sol (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":33.9009,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Sol (low)","score":33.9009,"secondary":{"configurationRank":85,"costPerTaskUsd":0.1326,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-sol-low","shortName":"GPT-6 Sol (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0.0909090909090909,"scicode":0.502314814814815,"hle":0.349397590361446},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:711bafc42ade2516","board":"aa_intelligence_index","model":"gpt-6-luna","variant":"GPT-6 Luna (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":33.8846,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Luna (xhigh)","score":33.8846,"secondary":{"configurationRank":86,"costPerTaskUsd":0.0421,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-luna-xhigh","shortName":"GPT-6 Luna (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"xhigh","terminalBench40":0.0808080808080808,"scicode":0.517361111111111,"hle":0.343373493975904},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:22a192a88d4be69a","board":"aa_intelligence_index","model":"gpt-5.5","variant":"GPT-5.5 (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":33.8044,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_intelligence_index","variantLabel":"GPT-5.5 (medium)","score":33.8044,"secondary":{"configurationRank":87,"costPerTaskUsd":0.9017,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-5-medium","shortName":"GPT-5.5 (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.0505050505050505,"scicode":0.545138888888889,"hle":0.42354031510658},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:4b9229f7a5210ab7","board":"aa_intelligence_index","model":"glm-5.2","variant":"GLM-5.2 (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":33.7055,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.2","boardId":"aa_intelligence_index","variantLabel":"GLM-5.2 (max)","score":33.7055,"secondary":{"configurationRank":89,"costPerTaskUsd":1.4726,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"glm-5-2","shortName":"GLM-5.2 (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"Z AI","effort":"max","terminalBench40":0.0101010101010101,"scicode":0.511574074074074,"hle":0.411492122335496},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:4555232c8e387389","board":"aa_intelligence_index","model":"qwen3.8-27b","variant":"Qwen3.8 27B (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":33.6963,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_intelligence_index","variantLabel":"Qwen3.8 27B (xhigh)","score":33.6963,"secondary":{"configurationRank":90,"costPerTaskUsd":1.0073,"scoreRoundedDisplay":34,"intelligenceIndexIsEstimated":false,"slug":"qwen3-8-27b","shortName":"Qwen3.8 27B (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"Alibaba","effort":"xhigh","terminalBench40":0.0555555555555556,"scicode":0.466435185185185,"hle":0.33920296570899},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:d1b16dee33968223","board":"aa_intelligence_index","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":33.4731,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Sol (low)","score":33.4731,"secondary":{"configurationRank":93,"costPerTaskUsd":0.2606,"scoreRoundedDisplay":33,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-sol-low","shortName":"GPT-5.6 Sol (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0.0101010101010101,"scicode":0.563657407407407,"hle":0.393883225208526},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:e0b41460917d3eff","board":"aa_intelligence_index","model":"gemini-3.8-flash","variant":"Gemini 3.8 Flash (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":33.4548,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.8 Flash (low)","score":33.4548,"secondary":{"configurationRank":94,"costPerTaskUsd":null,"scoreRoundedDisplay":33,"intelligenceIndexIsEstimated":false,"slug":"gemini-3-8-flash-low","shortName":"Gemini 3.8 Flash (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"Google","effort":"low","terminalBench40":0.101010101010101,"scicode":0.549768518518518,"hle":0.370713623725672},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:927f29d66e09bcc0","board":"aa_intelligence_index","model":"gpt-6-luna","variant":"GPT-6 Luna (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":32.1482,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Luna (high)","score":32.1482,"secondary":{"configurationRank":98,"costPerTaskUsd":0.0291,"scoreRoundedDisplay":32,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-luna-high","shortName":"GPT-6 Luna (high)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.0454545454545455,"scicode":0.503472222222222,"hle":0.329471733086191},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:384978b43afdd8f4","board":"aa_intelligence_index","model":"gpt-5.6-luna","variant":"GPT-5.6 Luna (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":32.1196,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Luna (high)","score":32.1196,"secondary":{"configurationRank":99,"costPerTaskUsd":0.044,"scoreRoundedDisplay":32,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-luna-high","shortName":"GPT-5.6 Luna (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"high","terminalBench40":0.0252525252525253,"scicode":0.516203703703704,"hle":0.334105653382762},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c089bab844997713","board":"aa_intelligence_index","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":31.6623,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5 (high)","score":31.6623,"secondary":{"configurationRank":101,"costPerTaskUsd":1.7925,"scoreRoundedDisplay":32,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-high","shortName":"Claude Sonnet 5 (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"high","terminalBench40":0.0505050505050505,"scicode":0.542824074074074,"hle":0.357275254865616},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c3e9b37ac894d3d3","board":"aa_intelligence_index","model":"gpt-5.5","variant":"GPT-5.5 (low) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":30.7095,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_intelligence_index","variantLabel":"GPT-5.5 (low)","score":30.7095,"secondary":{"configurationRank":104,"costPerTaskUsd":null,"scoreRoundedDisplay":31,"intelligenceIndexIsEstimated":true,"slug":"gpt-5-5-low","shortName":"GPT-5.5 (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":null,"scicode":null,"hle":0.326691380908248},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:99ded62ad6efb9a5","board":"aa_intelligence_index","model":"deepseek-v4-pro","variant":"DeepSeek V4 Pro (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":30.4497,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Pro (max)","score":30.4497,"secondary":{"configurationRank":106,"costPerTaskUsd":0.1215,"scoreRoundedDisplay":30,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-pro-0424","shortName":"DeepSeek V4 Pro (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"max","terminalBench40":0.146464646464646,"scicode":0.508101851851852,"hle":0.375347544022243},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:1635385ef12eb561","board":"aa_intelligence_index","model":"deepseek-v4-pro","variant":"DeepSeek V4 Pro (high) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":30.1055,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Pro (high)","score":30.1055,"secondary":{"configurationRank":108,"costPerTaskUsd":null,"scoreRoundedDisplay":30,"intelligenceIndexIsEstimated":true,"slug":"deepseek-v4-pro-0424-high","shortName":"DeepSeek V4 Pro (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"high","terminalBench40":null,"scicode":null,"hle":0.352177942539388},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:32f3726769b48fd3","board":"aa_intelligence_index","model":"gpt-5.6-terra","variant":"GPT-5.6 Terra (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":30.0938,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Terra (medium)","score":30.0938,"secondary":{"configurationRank":109,"costPerTaskUsd":0.1834,"scoreRoundedDisplay":30,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-terra-medium","shortName":"GPT-5.6 Terra (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.0101010101010101,"scicode":0.50462962962963,"hle":0.332715477293791},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:50501b171b1855d0","board":"aa_intelligence_index","model":"kimi-k3","variant":"Kimi K3 (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":30.0667,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k3","boardId":"aa_intelligence_index","variantLabel":"Kimi K3 (low)","score":30.0667,"secondary":{"configurationRank":110,"costPerTaskUsd":1.1463,"scoreRoundedDisplay":30,"intelligenceIndexIsEstimated":false,"slug":"kimi-k3-low","shortName":"Kimi K3 (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"Kimi","effort":"low","terminalBench40":0.126262626262626,"scicode":0.52662037037037,"hle":0.249768303985171},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:cc1fed45bce203af","board":"aa_intelligence_index","model":"gemini-3.1-pro","variant":"Gemini 3.1 Pro Preview","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":29.7186,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.1-pro","boardId":"aa_intelligence_index","variantLabel":"Gemini 3.1 Pro Preview","score":29.7186,"secondary":{"configurationRank":112,"costPerTaskUsd":0.6747,"scoreRoundedDisplay":30,"intelligenceIndexIsEstimated":false,"slug":"gemini-3-1-pro-preview","shortName":"Gemini 3.1 Pro Preview","isReasoning":true,"deprecated":false,"modelCreatorName":"Google","effort":null,"terminalBench40":0.0404040404040404,"scicode":0.586805555555556,"hle":0.470342910101946},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:51051c5b6fc81a75","board":"aa_intelligence_index","model":"gpt-6-luna","variant":"GPT-6 Luna (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":29.462,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Luna (medium)","score":29.462,"secondary":{"configurationRank":113,"costPerTaskUsd":0.0176,"scoreRoundedDisplay":29,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-luna-medium","shortName":"GPT-6 Luna (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.0252525252525253,"scicode":0.509259259259259,"hle":0.282669138090825},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b3246639fa93a057","board":"aa_intelligence_index","model":"qwen3.7-max","variant":"Qwen3.7 Max","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":29.4572,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.7-max","boardId":"aa_intelligence_index","variantLabel":"Qwen3.7 Max","score":29.4572,"secondary":{"configurationRank":114,"costPerTaskUsd":1.1473,"scoreRoundedDisplay":29,"intelligenceIndexIsEstimated":false,"slug":"qwen3-7-max","shortName":"Qwen3.7 Max","isReasoning":true,"deprecated":true,"modelCreatorName":"Alibaba","effort":"max","terminalBench40":0.0151515151515152,"scicode":0.49537037037037,"hle":0.405004633920297},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:179c62d659734cb1","board":"aa_intelligence_index","model":"minimax-m3","variant":"MiniMax-M3","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":29.2203,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"minimax-m3","boardId":"aa_intelligence_index","variantLabel":"MiniMax-M3","score":29.2203,"secondary":{"configurationRank":115,"costPerTaskUsd":0.5076,"scoreRoundedDisplay":29,"intelligenceIndexIsEstimated":false,"slug":"minimax-m3","shortName":"MiniMax-M3","isReasoning":true,"deprecated":false,"modelCreatorName":"MiniMax","effort":null,"terminalBench40":0.0202020202020202,"scicode":0.471064814814815,"hle":0.389712696941613},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:3a58445ae77253e4","board":"aa_intelligence_index","model":"qwen3.6-max-preview","variant":"Qwen3.6 Max Preview [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":28.3749,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.6-max-preview","boardId":"aa_intelligence_index","variantLabel":"Qwen3.6 Max Preview","score":28.3749,"secondary":{"configurationRank":119,"costPerTaskUsd":null,"scoreRoundedDisplay":28,"intelligenceIndexIsEstimated":true,"slug":"qwen3-6-max","shortName":"Qwen3.6 Max Preview","isReasoning":true,"deprecated":true,"modelCreatorName":"Alibaba","effort":"max","terminalBench40":null,"scicode":null,"hle":0.308155699721965},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:15f1554828f5b0dc","board":"aa_intelligence_index","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":28.328,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Sol (non-reasoning)","score":28.328,"secondary":{"configurationRank":120,"costPerTaskUsd":null,"scoreRoundedDisplay":28,"intelligenceIndexIsEstimated":true,"slug":"gpt-5-6-sol-non-reasoning","shortName":"GPT-5.6 Sol (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"OpenAI","effort":"non-reasoning","terminalBench40":null,"scicode":0.476851851851852,"hle":0.166821130676552},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:49ed8f07e3584753","board":"aa_intelligence_index","model":"gpt-6-sol","variant":"GPT-6 Sol (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":28.0922,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Sol (non-reasoning)","score":28.0922,"secondary":{"configurationRank":123,"costPerTaskUsd":0.3334,"scoreRoundedDisplay":28,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-sol-non-reasoning","shortName":"GPT-6 Sol (non-reasoning)","isReasoning":false,"deprecated":false,"modelCreatorName":"OpenAI","effort":"non-reasoning","terminalBench40":0.131313131313131,"scicode":0.47337962962963,"hle":0.183966635773865},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2b12f7c3563de350","board":"aa_intelligence_index","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":28.0531,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5 (medium)","score":28.0531,"secondary":{"configurationRank":124,"costPerTaskUsd":0.9991,"scoreRoundedDisplay":28,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-medium","shortName":"Claude Sonnet 5 (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"medium","terminalBench40":0.0202020202020202,"scicode":0.516203703703704,"hle":0.299814643188137},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b5ff5400e6b8641a","board":"aa_intelligence_index","model":"glm-5","variant":"GLM-5 [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":27.9112,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5","boardId":"aa_intelligence_index","variantLabel":"GLM-5","score":27.9112,"secondary":{"configurationRank":126,"costPerTaskUsd":null,"scoreRoundedDisplay":28,"intelligenceIndexIsEstimated":true,"slug":"glm-5","shortName":"GLM-5","isReasoning":true,"deprecated":true,"modelCreatorName":"Z AI","effort":null,"terminalBench40":null,"scicode":null,"hle":0.292863762743281},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:386e5826aefad337","board":"aa_intelligence_index","model":"qwen3.8-27b","variant":"Qwen3.8 27B (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":27.5508,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_intelligence_index","variantLabel":"Qwen3.8 27B (medium)","score":27.5508,"secondary":{"configurationRank":128,"costPerTaskUsd":1.1331,"scoreRoundedDisplay":28,"intelligenceIndexIsEstimated":false,"slug":"qwen3-8-27b-medium","shortName":"Qwen3.8 27B (medium)","isReasoning":true,"deprecated":false,"modelCreatorName":"Alibaba","effort":"medium","terminalBench40":0.0505050505050505,"scicode":0.390046296296296,"hle":0.140871177015755},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:fcef5db6f44a9168","board":"aa_intelligence_index","model":"gpt-5.6-terra","variant":"GPT-5.6 Terra (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":27.4962,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Terra (low)","score":27.4962,"secondary":{"configurationRank":129,"costPerTaskUsd":0.1445,"scoreRoundedDisplay":27,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-terra-low","shortName":"GPT-5.6 Terra (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0.0151515151515152,"scicode":0.498842592592593,"hle":0.291936978683967},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:238aba7e1b811926","board":"aa_intelligence_index","model":"qwen3.6-plus","variant":"Qwen3.6 Plus [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":27.0099,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.6-plus","boardId":"aa_intelligence_index","variantLabel":"Qwen3.6 Plus","score":27.0099,"secondary":{"configurationRank":132,"costPerTaskUsd":null,"scoreRoundedDisplay":27,"intelligenceIndexIsEstimated":true,"slug":"qwen3-6-plus","shortName":"Qwen3.6 Plus","isReasoning":true,"deprecated":true,"modelCreatorName":"Alibaba","effort":null,"terminalBench40":null,"scicode":null,"hle":0.278498609823911},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:97c970284455bffa","board":"aa_intelligence_index","model":"kimi-k2.6","variant":"Kimi K2.6","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":26.9792,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.6","boardId":"aa_intelligence_index","variantLabel":"Kimi K2.6","score":26.9792,"secondary":{"configurationRank":133,"costPerTaskUsd":0.7998,"scoreRoundedDisplay":27,"intelligenceIndexIsEstimated":false,"slug":"kimi-k2-6","shortName":"Kimi K2.6","isReasoning":true,"deprecated":true,"modelCreatorName":"Kimi","effort":null,"terminalBench40":0.00505050505050505,"scicode":0.515046296296296,"hle":0.374884151992586},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:d136741f62069cb1","board":"aa_intelligence_index","model":"qwen3.8-27b","variant":"Qwen3.8 27B (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":26.2048,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_intelligence_index","variantLabel":"Qwen3.8 27B (low)","score":26.2048,"secondary":{"configurationRank":140,"costPerTaskUsd":1.0482,"scoreRoundedDisplay":26,"intelligenceIndexIsEstimated":false,"slug":"qwen3-8-27b-low","shortName":"Qwen3.8 27B (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"Alibaba","effort":"low","terminalBench40":0.0252525252525253,"scicode":0.400462962962963,"hle":0.140407784986098},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:f45138a64ed521bd","board":"aa_intelligence_index","model":"glm-5.1","variant":"GLM-5.1","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":26.0586,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.1","boardId":"aa_intelligence_index","variantLabel":"GLM-5.1","score":26.0586,"secondary":{"configurationRank":141,"costPerTaskUsd":1.0193,"scoreRoundedDisplay":26,"intelligenceIndexIsEstimated":false,"slug":"glm-5-1","shortName":"GLM-5.1","isReasoning":true,"deprecated":true,"modelCreatorName":"Z AI","effort":null,"terminalBench40":0.0202020202020202,"scicode":0.447916666666667,"hle":0.300741427247451},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:c0536a98a4f2e51d","board":"aa_intelligence_index","model":"mimo-v2.5-pro","variant":"MiMo-V2.5-Pro","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":25.9869,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.5-pro","boardId":"aa_intelligence_index","variantLabel":"MiMo-V2.5-Pro","score":25.9869,"secondary":{"configurationRank":143,"costPerTaskUsd":0.0541,"scoreRoundedDisplay":26,"intelligenceIndexIsEstimated":false,"slug":"mimo-v2-5-pro","shortName":"MiMo-V2.5-Pro","isReasoning":true,"deprecated":true,"modelCreatorName":"Xiaomi","effort":null,"terminalBench40":0,"scicode":0.505787037037037,"hle":0.356811862835959},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b0c9eb24f253523d","board":"aa_intelligence_index","model":"kimi-k2.7-code","variant":"Kimi K2.7 Code","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":25.8121,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.7-code","boardId":"aa_intelligence_index","variantLabel":"Kimi K2.7 Code","score":25.8121,"secondary":{"configurationRank":144,"costPerTaskUsd":0.5412,"scoreRoundedDisplay":26,"intelligenceIndexIsEstimated":false,"slug":"kimi-k2-7-code","shortName":"Kimi K2.7 Code","isReasoning":true,"deprecated":false,"modelCreatorName":"Kimi","effort":null,"terminalBench40":0.0101010101010101,"scicode":0.478009259259259,"hle":0.35032437442076},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:2ecad05994d7b1af","board":"aa_intelligence_index","model":"inkling-small","variant":"Inkling Small","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":25.6583,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"inkling-small","boardId":"aa_intelligence_index","variantLabel":"Inkling Small","score":25.6583,"secondary":{"configurationRank":145,"costPerTaskUsd":0.0895,"scoreRoundedDisplay":26,"intelligenceIndexIsEstimated":false,"slug":"inkling-small","shortName":"Inkling Small","isReasoning":true,"deprecated":false,"modelCreatorName":"Thinking Machines","effort":null,"terminalBench40":0.0101010101010101,"scicode":0.496527777777778,"hle":0.333178869323448},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:131d42e954c0324c","board":"aa_intelligence_index","model":"hy3","variant":"Hy3","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":25.2973,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"hy3","boardId":"aa_intelligence_index","variantLabel":"Hy3","score":25.2973,"secondary":{"configurationRank":148,"costPerTaskUsd":0.0718,"scoreRoundedDisplay":25,"intelligenceIndexIsEstimated":false,"slug":"hy3","shortName":"Hy3","isReasoning":true,"deprecated":false,"modelCreatorName":"Tencent","effort":null,"terminalBench40":0.00505050505050505,"scicode":0.486111111111111,"hle":0.334569045412419},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:4f0191d02e894561","board":"aa_intelligence_index","model":"mimo-v2.5","variant":"MiMo-V2.5 [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":25.1743,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.5","boardId":"aa_intelligence_index","variantLabel":"MiMo-V2.5","score":25.1743,"secondary":{"configurationRank":150,"costPerTaskUsd":null,"scoreRoundedDisplay":25,"intelligenceIndexIsEstimated":true,"slug":"mimo-v2-5-0424","shortName":"MiMo-V2.5","isReasoning":true,"deprecated":true,"modelCreatorName":"Xiaomi","effort":null,"terminalBench40":0,"scicode":0.438657407407407,"hle":0.272011121408712},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:2e94df38c6b223ca","board":"aa_intelligence_index","model":"qwen3.7-plus","variant":"Qwen3.7 Plus","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":25.1622,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.7-plus","boardId":"aa_intelligence_index","variantLabel":"Qwen3.7 Plus","score":25.1622,"secondary":{"configurationRank":151,"costPerTaskUsd":0.217,"scoreRoundedDisplay":25,"intelligenceIndexIsEstimated":false,"slug":"qwen3-7-plus","shortName":"Qwen3.7 Plus","isReasoning":true,"deprecated":false,"modelCreatorName":"Alibaba","effort":null,"terminalBench40":0.0101010101010101,"scicode":0.460648148148148,"hle":0.356348470806302},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:60904975d230578f","board":"aa_intelligence_index","model":"gpt-5.6-luna","variant":"GPT-5.6 Luna (medium)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":25.0359,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Luna (medium)","score":25.0359,"secondary":{"configurationRank":153,"costPerTaskUsd":0.0156,"scoreRoundedDisplay":25,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-luna-medium","shortName":"GPT-5.6 Luna (medium)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"medium","terminalBench40":0.00505050505050505,"scicode":0.467592592592593,"hle":0.257645968489342},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:ae97bbe112769bcd","board":"aa_intelligence_index","model":"inkling","variant":"Inkling (xhigh)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":24.9848,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"inkling","boardId":"aa_intelligence_index","variantLabel":"Inkling (xhigh)","score":24.9848,"secondary":{"configurationRank":154,"costPerTaskUsd":null,"scoreRoundedDisplay":25,"intelligenceIndexIsEstimated":false,"slug":"inkling","shortName":"Inkling (xhigh)","isReasoning":true,"deprecated":false,"modelCreatorName":"Thinking Machines","effort":"xhigh","terminalBench40":0.0101010101010101,"scicode":0.469907407407407,"hle":0.318813716404078},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:d81376edf86feffa","board":"aa_intelligence_index","model":"deepseek-v4.1-flash","variant":"DeepSeek V4.1 Flash (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":24.6734,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4.1-flash","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4.1 Flash (non-reasoning)","score":24.6734,"secondary":{"configurationRank":161,"costPerTaskUsd":0.1467,"scoreRoundedDisplay":25,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-1-flash-non-reasoning","shortName":"DeepSeek V4.1 Flash (non-reasoning)","isReasoning":false,"deprecated":false,"modelCreatorName":"DeepSeek","effort":"non-reasoning","terminalBench40":0.0555555555555556,"scicode":0.355324074074074,"hle":0.107970342910102},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:7cb5e674535a254b","board":"aa_intelligence_index","model":"deepseek-v4-flash","variant":"DeepSeek V4 Flash (high)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":24.384,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Flash (high)","score":24.384,"secondary":{"configurationRank":163,"costPerTaskUsd":0.1271,"scoreRoundedDisplay":24,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-flash-0420-high","shortName":"DeepSeek V4 Flash (high)","isReasoning":true,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"high","terminalBench40":0.0303030303030303,"scicode":0.40162037037037,"hle":0.30259499536608},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:b99819b727eb71df","board":"aa_intelligence_index","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":24.2638,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5 (low)","score":24.2638,"secondary":{"configurationRank":165,"costPerTaskUsd":0.5088,"scoreRoundedDisplay":24,"intelligenceIndexIsEstimated":false,"slug":"claude-sonnet-5-low","shortName":"Claude Sonnet 5 (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"Anthropic","effort":"low","terminalBench40":0.0252525252525253,"scicode":0.501157407407407,"hle":0.219184430027804},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:18011484156b18b6","board":"aa_intelligence_index","model":"glm-5.1","variant":"GLM-5.1 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":24.2412,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.1","boardId":"aa_intelligence_index","variantLabel":"GLM-5.1 (non-reasoning)","score":24.2412,"secondary":{"configurationRank":166,"costPerTaskUsd":null,"scoreRoundedDisplay":24,"intelligenceIndexIsEstimated":true,"slug":"glm-5-1-non-reasoning","shortName":"GLM-5.1 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Z AI","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.279425393883225},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:db4df116f5cc1c83","board":"aa_intelligence_index","model":"deepseek-v4-flash","variant":"DeepSeek V4 Flash (max)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":24.1673,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Flash (max)","score":24.1673,"secondary":{"configurationRank":167,"costPerTaskUsd":0.1075,"scoreRoundedDisplay":24,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-flash-0420","shortName":"DeepSeek V4 Flash (max)","isReasoning":true,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"max","terminalBench40":0.0252525252525253,"scicode":0.452546296296296,"hle":0.348470806302132},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:be58b1c3ec11fb7d","board":"aa_intelligence_index","model":"kimi-k2.6","variant":"Kimi K2.6 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":23.5696,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.6","boardId":"aa_intelligence_index","variantLabel":"Kimi K2.6 (non-reasoning)","score":23.5696,"secondary":{"configurationRank":174,"costPerTaskUsd":null,"scoreRoundedDisplay":24,"intelligenceIndexIsEstimated":true,"slug":"kimi-k2-6-non-reasoning","shortName":"Kimi K2.6 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Kimi","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.195551436515292},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:589e83427c9b8ab7","board":"aa_intelligence_index","model":"kimi-k2.5","variant":"Kimi K2.5 [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":23.4585,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.5","boardId":"aa_intelligence_index","variantLabel":"Kimi K2.5","score":23.4585,"secondary":{"configurationRank":176,"costPerTaskUsd":null,"scoreRoundedDisplay":23,"intelligenceIndexIsEstimated":true,"slug":"kimi-k2-5","shortName":"Kimi K2.5","isReasoning":true,"deprecated":true,"modelCreatorName":"Kimi","effort":null,"terminalBench40":null,"scicode":null,"hle":0.307228915662651},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:e35c69aedd168116","board":"aa_intelligence_index","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":23.2,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_intelligence_index","variantLabel":"Claude Sonnet 5 (non-reasoning)","score":23.2,"secondary":{"configurationRank":178,"costPerTaskUsd":null,"scoreRoundedDisplay":23,"intelligenceIndexIsEstimated":true,"slug":"claude-sonnet-5-non-reasoning","shortName":"Claude Sonnet 5 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Anthropic","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.190454124189064},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:e75792c6ebb8bb80","board":"aa_intelligence_index","model":"gpt-5.5","variant":"GPT-5.5 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":23.1674,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_intelligence_index","variantLabel":"GPT-5.5 (non-reasoning)","score":23.1674,"secondary":{"configurationRank":179,"costPerTaskUsd":null,"scoreRoundedDisplay":23,"intelligenceIndexIsEstimated":true,"slug":"gpt-5-5-non-reasoning","shortName":"GPT-5.5 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"OpenAI","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.137164040778499},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:5cd72513bff7d4b7","board":"aa_intelligence_index","model":"nemotron-3-ultra","variant":"Nemotron 3 Ultra","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":22.9279,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"nemotron-3-ultra","boardId":"aa_intelligence_index","variantLabel":"Nemotron 3 Ultra","score":22.9279,"secondary":{"configurationRank":181,"costPerTaskUsd":0.5976,"scoreRoundedDisplay":23,"intelligenceIndexIsEstimated":false,"slug":"nvidia-nemotron-3-ultra-550b-a55b","shortName":"Nemotron 3 Ultra","isReasoning":true,"deprecated":false,"modelCreatorName":"NVIDIA","effort":null,"terminalBench40":0.00505050505050505,"scicode":0.402777777777778,"hle":0.284059314179796},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:ae390fdc5fe89d6c","board":"aa_intelligence_index","model":"minimax-m2.5","variant":"MiniMax-M2.5 [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":22.7966,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"minimax-m2.5","boardId":"aa_intelligence_index","variantLabel":"MiniMax-M2.5","score":22.7966,"secondary":{"configurationRank":185,"costPerTaskUsd":null,"scoreRoundedDisplay":23,"intelligenceIndexIsEstimated":true,"slug":"minimax-m2-5","shortName":"MiniMax-M2.5","isReasoning":true,"deprecated":true,"modelCreatorName":"MiniMax","effort":null,"terminalBench40":null,"scicode":null,"hle":0.205282669138091},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:58eba7a1df2c56c5","board":"aa_intelligence_index","model":"minimax-m2.7","variant":"MiniMax-M2.7","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":22.7578,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"minimax-m2.7","boardId":"aa_intelligence_index","variantLabel":"MiniMax-M2.7","score":22.7578,"secondary":{"configurationRank":186,"costPerTaskUsd":null,"scoreRoundedDisplay":23,"intelligenceIndexIsEstimated":false,"slug":"minimax-m2-7","shortName":"MiniMax-M2.7","isReasoning":true,"deprecated":true,"modelCreatorName":"MiniMax","effort":null,"terminalBench40":0,"scicode":0.501157407407407,"hle":0.29610750695088},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:bf6c2c604a56c9d0","board":"aa_intelligence_index","model":"glm-5.2","variant":"GLM-5.2 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":22.4323,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.2","boardId":"aa_intelligence_index","variantLabel":"GLM-5.2 (non-reasoning)","score":22.4323,"secondary":{"configurationRank":193,"costPerTaskUsd":null,"scoreRoundedDisplay":22,"intelligenceIndexIsEstimated":true,"slug":"glm-5-2-non-reasoning","shortName":"GLM-5.2 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Z AI","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.097775718257646},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:81bcbed3fd5a66b9","board":"aa_intelligence_index","model":"glm-5","variant":"GLM-5 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":21.7826,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5","boardId":"aa_intelligence_index","variantLabel":"GLM-5 (non-reasoning)","score":21.7826,"secondary":{"configurationRank":200,"costPerTaskUsd":null,"scoreRoundedDisplay":22,"intelligenceIndexIsEstimated":true,"slug":"glm-5-non-reasoning","shortName":"GLM-5 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Z AI","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.0764596848934198},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:5a5591e8d01f6bcc","board":"aa_intelligence_index","model":"gpt-5.6-luna","variant":"GPT-5.6 Luna (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":21.0125,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Luna (low)","score":21.0125,"secondary":{"configurationRank":206,"costPerTaskUsd":0.0098,"scoreRoundedDisplay":21,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-luna-low","shortName":"GPT-5.6 Luna (low)","isReasoning":true,"deprecated":true,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0,"scicode":0.460648148148148,"hle":0.198331788693234},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:9e4a1080bc5cdeda","board":"aa_intelligence_index","model":"gpt-6-luna","variant":"GPT-6 Luna (low)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"low","service_mode":null,"score":20.9225,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Luna (low)","score":20.9225,"secondary":{"configurationRank":208,"costPerTaskUsd":0.0045,"scoreRoundedDisplay":21,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-luna-low","shortName":"GPT-6 Luna (low)","isReasoning":true,"deprecated":false,"modelCreatorName":"OpenAI","effort":"low","terminalBench40":0,"scicode":0.46875,"hle":0.202502316960148},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:e15aa3daea8ed086","board":"aa_intelligence_index","model":"deepseek-v4-pro","variant":"DeepSeek V4 Pro (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":20.8345,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Pro (non-reasoning)","score":20.8345,"secondary":{"configurationRank":209,"costPerTaskUsd":null,"scoreRoundedDisplay":21,"intelligenceIndexIsEstimated":true,"slug":"deepseek-v4-pro-0424-non-reasoning","shortName":"DeepSeek V4 Pro (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.082483781278962},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:d3febc87f28b66bc","board":"aa_intelligence_index","model":"gpt-5.6-terra","variant":"GPT-5.6 Terra (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":20.7797,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Terra (non-reasoning)","score":20.7797,"secondary":{"configurationRank":212,"costPerTaskUsd":0.1397,"scoreRoundedDisplay":21,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-terra-non-reasoning","shortName":"GPT-5.6 Terra (non-reasoning)","isReasoning":false,"deprecated":false,"modelCreatorName":"OpenAI","effort":"non-reasoning","terminalBench40":0.00505050505050505,"scicode":0.451388888888889,"hle":0.113994439295644},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:f323daf280bfff22","board":"aa_intelligence_index","model":"deepseek-v4-pro","variant":"DeepSeek V4 Pro 0813 (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":20.3605,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Pro 0813 (non-reasoning)","score":20.3605,"secondary":{"configurationRank":221,"costPerTaskUsd":0.4774,"scoreRoundedDisplay":20,"intelligenceIndexIsEstimated":false,"slug":"deepseek-v4-pro-non-reasoning","shortName":"DeepSeek V4 Pro 0813 (non-reasoning)","isReasoning":false,"deprecated":false,"modelCreatorName":"DeepSeek","effort":"non-reasoning","terminalBench40":0.0151515151515152,"scicode":0.399305555555556,"hle":0.105653382761817},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:7bed7be52e30496d","board":"aa_intelligence_index","model":"qwen3.8-27b","variant":"Qwen3.8 27B (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":20.1502,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_intelligence_index","variantLabel":"Qwen3.8 27B (non-reasoning)","score":20.1502,"secondary":{"configurationRank":223,"costPerTaskUsd":2.4878,"scoreRoundedDisplay":20,"intelligenceIndexIsEstimated":false,"slug":"qwen3-8-27b-non-reasoning","shortName":"Qwen3.8 27B (non-reasoning)","isReasoning":false,"deprecated":false,"modelCreatorName":"Alibaba","effort":"non-reasoning","terminalBench40":0,"scicode":0.362268518518519,"hle":0.121408711770158},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:e8828ee245d130af","board":"aa_intelligence_index","model":"step-3.7-flash","variant":"Step 3.7 Flash [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":19.4813,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"step-3.7-flash","boardId":"aa_intelligence_index","variantLabel":"Step 3.7 Flash","score":19.4813,"secondary":{"configurationRank":229,"costPerTaskUsd":null,"scoreRoundedDisplay":19,"intelligenceIndexIsEstimated":true,"slug":"step-3-7-flash","shortName":"Step 3.7 Flash","isReasoning":true,"deprecated":true,"modelCreatorName":"StepFun","effort":null,"terminalBench40":null,"scicode":0.438657407407407,"hle":0.214087117701576},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:1afe01378404f2e7","board":"aa_intelligence_index","model":"kimi-k2.5","variant":"Kimi K2.5 (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":19.4343,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.5","boardId":"aa_intelligence_index","variantLabel":"Kimi K2.5 (non-reasoning)","score":19.4343,"secondary":{"configurationRank":230,"costPerTaskUsd":null,"scoreRoundedDisplay":19,"intelligenceIndexIsEstimated":true,"slug":"kimi-k2-5-non-reasoning","shortName":"Kimi K2.5 (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Kimi","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.131603336422614},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:73481783e67e2f5b","board":"aa_intelligence_index","model":"longcat-2.0","variant":"LongCat 2.0","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":19.1123,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"longcat-2.0","boardId":"aa_intelligence_index","variantLabel":"LongCat 2.0","score":19.1123,"secondary":{"configurationRank":234,"costPerTaskUsd":0.059,"scoreRoundedDisplay":19,"intelligenceIndexIsEstimated":false,"slug":"longcat-2-0","shortName":"LongCat 2.0","isReasoning":true,"deprecated":false,"modelCreatorName":"LongCat","effort":null,"terminalBench40":0,"scicode":0.363425925925926,"hle":0.336886005560704},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:610f4083dbb622ce","board":"aa_intelligence_index","model":"deepseek-v4-flash","variant":"DeepSeek V4 Flash (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":18.8769,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_intelligence_index","variantLabel":"DeepSeek V4 Flash (non-reasoning)","score":18.8769,"secondary":{"configurationRank":236,"costPerTaskUsd":null,"scoreRoundedDisplay":19,"intelligenceIndexIsEstimated":true,"slug":"deepseek-v4-flash-0420-non-reasoning","shortName":"DeepSeek V4 Flash (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"DeepSeek","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.0778498609823911},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:95331afc94452f27","board":"aa_intelligence_index","model":"mimo-v2.5-pro","variant":"MiMo-V2.5-Pro (non-reasoning) [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":18.29,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.5-pro","boardId":"aa_intelligence_index","variantLabel":"MiMo-V2.5-Pro (non-reasoning)","score":18.29,"secondary":{"configurationRank":243,"costPerTaskUsd":null,"scoreRoundedDisplay":18,"intelligenceIndexIsEstimated":true,"slug":"mimo-v2-5-pro-non-reasoning","shortName":"MiMo-V2.5-Pro (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"Xiaomi","effort":"non-reasoning","terminalBench40":null,"scicode":null,"hle":0.147822057460612},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:be701ec71aacde7a","board":"aa_intelligence_index","model":"gpt-6-luna","variant":"GPT-6 Luna (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":18.2617,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-6 Luna (non-reasoning)","score":18.2617,"secondary":{"configurationRank":244,"costPerTaskUsd":0.0115,"scoreRoundedDisplay":18,"intelligenceIndexIsEstimated":false,"slug":"gpt-6-luna-non-reasoning","shortName":"GPT-6 Luna (non-reasoning)","isReasoning":false,"deprecated":false,"modelCreatorName":"OpenAI","effort":"non-reasoning","terminalBench40":0.0151515151515152,"scicode":0.430555555555556,"hle":0.0857275254865616},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:f4ecc0cb880e6a93","board":"aa_intelligence_index","model":"step-3.5-flash","variant":"Step 3.5 Flash 2603 [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":16.9641,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"step-3.5-flash","boardId":"aa_intelligence_index","variantLabel":"Step 3.5 Flash 2603","score":16.9641,"secondary":{"configurationRank":256,"costPerTaskUsd":null,"scoreRoundedDisplay":17,"intelligenceIndexIsEstimated":true,"slug":"step-3-5-flash","shortName":"Step 3.5 Flash 2603","isReasoning":true,"deprecated":true,"modelCreatorName":"StepFun","effort":null,"terminalBench40":null,"scicode":null,"hle":0.244670991658943},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:19396846e36b1f6e","board":"aa_intelligence_index","model":"step-3.5-flash","variant":"Step 3.5 Flash [AA estimate]","score_is_estimated":true,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":16.6169,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"step-3.5-flash","boardId":"aa_intelligence_index","variantLabel":"Step 3.5 Flash","score":16.6169,"secondary":{"configurationRank":263,"costPerTaskUsd":null,"scoreRoundedDisplay":17,"intelligenceIndexIsEstimated":true,"slug":"step-3-5-flash-0202","shortName":"Step 3.5 Flash","isReasoning":true,"deprecated":true,"modelCreatorName":"StepFun","effort":null,"terminalBench40":null,"scicode":null,"hle":0.211306765523633},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured."}},{"configuration_id":"aa_intelligence_index:ee543e6460c00da2","board":"aa_intelligence_index","model":"gpt-5.6-luna","variant":"GPT-5.6 Luna (non-reasoning)","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":15.527,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_intelligence_index","variantLabel":"GPT-5.6 Luna (non-reasoning)","score":15.527,"secondary":{"configurationRank":275,"costPerTaskUsd":0.0101,"scoreRoundedDisplay":16,"intelligenceIndexIsEstimated":false,"slug":"gpt-5-6-luna-non-reasoning","shortName":"GPT-5.6 Luna (non-reasoning)","isReasoning":false,"deprecated":true,"modelCreatorName":"OpenAI","effort":"non-reasoning","terminalBench40":0.0101010101010101,"scicode":0.403935185185185,"hle":0.072289156626506},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_intelligence_index:479a072abc28cf2a","board":"aa_intelligence_index","model":"mistral-medium-3.5","variant":"Mistral Medium 3.5","score_is_estimated":false,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":14.1889,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mistral-medium-3.5","boardId":"aa_intelligence_index","variantLabel":"Mistral Medium 3.5","score":14.1889,"secondary":{"configurationRank":296,"costPerTaskUsd":0.502,"scoreRoundedDisplay":14,"intelligenceIndexIsEstimated":false,"slug":"mistral-medium-3-5","shortName":"Mistral Medium 3.5","isReasoning":true,"deprecated":false,"modelCreatorName":"Mistral","effort":null,"terminalBench40":0,"scicode":0.40162037037037,"hle":0.137627432808156},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01"}},{"configuration_id":"aa_terminal_bench_4:8cbc13578cabcb6a","board":"aa_terminal_bench_4","model":"claude-sonnet-5.5","variant":"Artificial Analysis - Claude Sonnet 5.5 (max with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":63.6364,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5.5 (max with fallback)","score":63.6364,"secondary":{"configurationRank":1,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"claude-sonnet-5-5","shortName":"Claude Sonnet 5.5 (max with fallback)","terminalBench40Fraction":0.636363636363636},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:700453154be041cb","board":"aa_terminal_bench_4","model":"gpt-6-astra","variant":"Artificial Analysis - GPT-6 Astra (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":59.596,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Astra (xhigh)","score":59.596,"secondary":{"configurationRank":2,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-6-astra-xhigh","shortName":"GPT-6 Astra (xhigh)","terminalBench40Fraction":0.595959595959596},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:2a264ae9483b05c3","board":"aa_terminal_bench_4","model":"claude-opus-5.5","variant":"Artificial Analysis - Claude Opus 5.5 (max with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":59.596,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5.5 (max with fallback)","score":59.596,"secondary":{"configurationRank":3,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"claude-opus-5-5","shortName":"Claude Opus 5.5 (max with fallback)","terminalBench40Fraction":0.595959595959596},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:23b9058c23d1ec26","board":"aa_terminal_bench_4","model":"claude-opus-5.5","variant":"Artificial Analysis - Claude Opus 5.5 (xhigh with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":59.596,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5.5 (xhigh with fallback)","score":59.596,"secondary":{"configurationRank":4,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"claude-opus-5-5-xhigh","shortName":"Claude Opus 5.5 (xhigh with fallback)","terminalBench40Fraction":0.595959595959596},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:e44812025afd889c","board":"aa_terminal_bench_4","model":"gpt-6-astra","variant":"Artificial Analysis - GPT-6 Astra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":59.0909,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Astra (max)","score":59.0909,"secondary":{"configurationRank":5,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-6-astra","shortName":"GPT-6 Astra (max)","terminalBench40Fraction":0.590909090909091},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:e58d9ba7080efc8c","board":"aa_terminal_bench_4","model":"claude-sonnet-5.5","variant":"Artificial Analysis - Claude Sonnet 5.5 (xhigh with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":57.0707,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5.5 (xhigh with fallback)","score":57.0707,"secondary":{"configurationRank":6,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"claude-sonnet-5-5-xhigh","shortName":"Claude Sonnet 5.5 (xhigh with fallback)","terminalBench40Fraction":0.570707070707071},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:2a1794e2e04395f3","board":"aa_terminal_bench_4","model":"claude-opus-5.5","variant":"Artificial Analysis - Claude Opus 5.5 (high with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":56.5657,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5.5 (high with fallback)","score":56.5657,"secondary":{"configurationRank":8,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"claude-opus-5-5-high","shortName":"Claude Opus 5.5 (high with fallback)","terminalBench40Fraction":0.565656565656566},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:f935d95ba064b29b","board":"aa_terminal_bench_4","model":"gpt-6.1-sol","variant":"Artificial Analysis - GPT-6.1 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":56.0606,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6.1 Sol (max)","score":56.0606,"secondary":{"configurationRank":9,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-6-1-sol","shortName":"GPT-6.1 Sol (max)","terminalBench40Fraction":0.560606060606061},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:41e12ce515ec9121","board":"aa_terminal_bench_4","model":"claude-fable-5.1","variant":"Artificial Analysis - Claude Fable 5.1 (xhigh with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":55.0505,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Fable 5.1 (xhigh with fallback)","score":55.0505,"secondary":{"configurationRank":10,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"claude-fable-5-1-xhigh","shortName":"Claude Fable 5.1 (xhigh with fallback)","terminalBench40Fraction":0.55050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:94af356c207fc48e","board":"aa_terminal_bench_4","model":"gpt-6.1-sol","variant":"Artificial Analysis - GPT-6.1 Sol (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":54.0404,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6.1 Sol (xhigh)","score":54.0404,"secondary":{"configurationRank":11,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-6-1-sol-xhigh","shortName":"GPT-6.1 Sol (xhigh)","terminalBench40Fraction":0.54040404040404},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:12ee7971061ca0e0","board":"aa_terminal_bench_4","model":"gpt-6-astra","variant":"Artificial Analysis - GPT-6 Astra (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":54.0404,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Astra (high)","score":54.0404,"secondary":{"configurationRank":12,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-6-astra-high","shortName":"GPT-6 Astra (high)","terminalBench40Fraction":0.54040404040404},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:21a53e51c76d68e5","board":"aa_terminal_bench_4","model":"claude-opus-5.5","variant":"Artificial Analysis - Claude Opus 5.5 (medium with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":52.5253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5.5 (medium with fallback)","score":52.5253,"secondary":{"configurationRank":13,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"claude-opus-5-5-medium","shortName":"Claude Opus 5.5 (medium with fallback)","terminalBench40Fraction":0.525252525252525},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:1512d68b0eb6bc1b","board":"aa_terminal_bench_4","model":"claude-fable-5.1","variant":"Artificial Analysis - Claude Fable 5.1 (max with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":52.0202,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Fable 5.1 (max with fallback)","score":52.0202,"secondary":{"configurationRank":14,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"claude-fable-5-1","shortName":"Claude Fable 5.1 (max with fallback)","terminalBench40Fraction":0.52020202020202},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:e606b01ceda846f9","board":"aa_terminal_bench_4","model":"claude-fable-5.1","variant":"Artificial Analysis - Claude Fable 5.1 (high with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":52.0202,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Fable 5.1 (high with fallback)","score":52.0202,"secondary":{"configurationRank":15,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"claude-fable-5-1-high","shortName":"Claude Fable 5.1 (high with fallback)","terminalBench40Fraction":0.52020202020202},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:c77efb6f723a47b3","board":"aa_terminal_bench_4","model":"gpt-6.1-sol","variant":"Artificial Analysis - GPT-6.1 Sol (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":51.5152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6.1 Sol (high)","score":51.5152,"secondary":{"configurationRank":16,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-6-1-sol-high","shortName":"GPT-6.1 Sol (high)","terminalBench40Fraction":0.515151515151515},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0634c4abcbad70cc","board":"aa_terminal_bench_4","model":"gpt-6-astra","variant":"Artificial Analysis - GPT-6 Astra (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":49.4949,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Astra (medium)","score":49.4949,"secondary":{"configurationRank":17,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-6-astra-medium","shortName":"GPT-6 Astra (medium)","terminalBench40Fraction":0.494949494949495},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0ed48e92962897a4","board":"aa_terminal_bench_4","model":"claude-opus-5","variant":"Artificial Analysis - Claude Opus 5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":48.9899,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5 (max)","score":48.9899,"secondary":{"configurationRank":18,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"claude-opus-5","shortName":"Claude Opus 5 (max)","terminalBench40Fraction":0.48989898989899},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:8f793b0822f5faf4","board":"aa_terminal_bench_4","model":"gpt-6.1-sol","variant":"Artificial Analysis - GPT-6.1 Sol (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":47.9798,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6.1 Sol (medium)","score":47.9798,"secondary":{"configurationRank":19,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-6-1-sol-medium","shortName":"GPT-6.1 Sol (medium)","terminalBench40Fraction":0.47979797979798},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0838ca37e5fa42da","board":"aa_terminal_bench_4","model":"claude-opus-5","variant":"Artificial Analysis - Claude Opus 5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":46.4646,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5 (xhigh)","score":46.4646,"secondary":{"configurationRank":20,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"claude-opus-5-xhigh","shortName":"Claude Opus 5 (xhigh)","terminalBench40Fraction":0.464646464646465},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:8e10956f03aaa14d","board":"aa_terminal_bench_4","model":"claude-opus-5","variant":"Artificial Analysis - Claude Opus 5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":45.9596,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5 (high)","score":45.9596,"secondary":{"configurationRank":21,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"claude-opus-5-high","shortName":"Claude Opus 5 (high)","terminalBench40Fraction":0.45959595959596},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:93ccfc6e6c1f4910","board":"aa_terminal_bench_4","model":"claude-fable-5.1","variant":"Artificial Analysis - Claude Fable 5.1 (medium with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":44.9495,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Fable 5.1 (medium with fallback)","score":44.9495,"secondary":{"configurationRank":22,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"claude-fable-5-1-medium","shortName":"Claude Fable 5.1 (medium with fallback)","terminalBench40Fraction":0.44949494949495},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:8571a16baadf9b9a","board":"aa_terminal_bench_4","model":"claude-sonnet-5.5","variant":"Artificial Analysis - Claude Sonnet 5.5 (high with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":43.9394,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5.5 (high with fallback)","score":43.9394,"secondary":{"configurationRank":23,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"claude-sonnet-5-5-high","shortName":"Claude Sonnet 5.5 (high with fallback)","terminalBench40Fraction":0.439393939393939},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:c54c846fa56e7fb9","board":"aa_terminal_bench_4","model":"gpt-6-sol","variant":"Artificial Analysis - GPT-6 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":43.9394,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Sol (max)","score":43.9394,"secondary":{"configurationRank":24,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-6-sol","shortName":"GPT-6 Sol (max)","terminalBench40Fraction":0.439393939393939},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:6fd65552e1fc1e0a","board":"aa_terminal_bench_4","model":"claude-fable-5","variant":"Artificial Analysis - Claude Fable 5 (with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":42.4242,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Fable 5 (with fallback)","score":42.4242,"secondary":{"configurationRank":25,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"claude-fable-5","shortName":"Claude Fable 5 (with fallback)","terminalBench40Fraction":0.424242424242424},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:4785a3610e775817","board":"aa_terminal_bench_4","model":"gpt-6-astra","variant":"Artificial Analysis - GPT-6 Astra (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":41.9192,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Astra (low)","score":41.9192,"secondary":{"configurationRank":26,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-6-astra-low","shortName":"GPT-6 Astra (low)","terminalBench40Fraction":0.419191919191919},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:496fda38cce0cdb6","board":"aa_terminal_bench_4","model":"glm-5.3","variant":"Artificial Analysis - GLM-5.3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":41.9192,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GLM-5.3 (max)","score":41.9192,"secondary":{"configurationRank":27,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"glm-5-3","shortName":"GLM-5.3 (max)","terminalBench40Fraction":0.419191919191919},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:5e01b08628a88aac","board":"aa_terminal_bench_4","model":"claude-fable-5.1","variant":"Artificial Analysis - Claude Fable 5.1 (low with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":40.404,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Fable 5.1 (low with fallback)","score":40.404,"secondary":{"configurationRank":28,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"claude-fable-5-1-low","shortName":"Claude Fable 5.1 (low with fallback)","terminalBench40Fraction":0.404040404040404},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:7ac4c013fed0b565","board":"aa_terminal_bench_4","model":"gpt-5.6-sol","variant":"Artificial Analysis - GPT-5.6 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":39.899,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Sol (max)","score":39.899,"secondary":{"configurationRank":29,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-5-6-sol","shortName":"GPT-5.6 Sol (max)","terminalBench40Fraction":0.398989898989899},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:696204be45eb44af","board":"aa_terminal_bench_4","model":"qwen3.8-max-0902","variant":"Artificial Analysis - Qwen3.8 Max (0902)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":38.8889,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-max-0902","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.8 Max (0902)","score":38.8889,"secondary":{"configurationRank":30,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"qwen3-8-max","shortName":"Qwen3.8 Max (0902)","terminalBench40Fraction":0.388888888888889},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:4ce13a3cd68939ad","board":"aa_terminal_bench_4","model":"gpt-5.6-terra","variant":"Artificial Analysis - GPT-5.6 Terra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":35.3535,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Terra (max)","score":35.3535,"secondary":{"configurationRank":31,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-5-6-terra","shortName":"GPT-5.6 Terra (max)","terminalBench40Fraction":0.353535353535354},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:b5691a8df0db5041","board":"aa_terminal_bench_4","model":"mimo-v2.6-pro","variant":"Artificial Analysis - MiMo-V2.6-Pro","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":34.8485,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.6-pro","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - MiMo-V2.6-Pro","score":34.8485,"secondary":{"configurationRank":32,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"mimo-v2-6-pro","shortName":"MiMo-V2.6-Pro","terminalBench40Fraction":0.348484848484849},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:69eafd0d76477763","board":"aa_terminal_bench_4","model":"glm-5.3","variant":"Artificial Analysis - GLM-5.3 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":34.8485,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GLM-5.3 (low)","score":34.8485,"secondary":{"configurationRank":33,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"glm-5-3-low","shortName":"GLM-5.3 (low)","terminalBench40Fraction":0.348484848484849},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:314fe657e048e298","board":"aa_terminal_bench_4","model":"claude-opus-5","variant":"Artificial Analysis - Claude Opus 5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":34.3434,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5 (medium)","score":34.3434,"secondary":{"configurationRank":34,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"claude-opus-5-medium","shortName":"Claude Opus 5 (medium)","terminalBench40Fraction":0.343434343434343},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:e7fa06d82cc14898","board":"aa_terminal_bench_4","model":"step-5-preview","variant":"Artificial Analysis - Step 5 Preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":33.3333,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"step-5-preview","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Step 5 Preview","score":33.3333,"secondary":{"configurationRank":35,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"step-5","shortName":"Step 5 Preview","terminalBench40Fraction":0.333333333333333},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:3ceb10cebdd638c2","board":"aa_terminal_bench_4","model":"muse-spark-1.3","variant":"Artificial Analysis - Muse Spark 1.3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":33.3333,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Muse Spark 1.3 (max)","score":33.3333,"secondary":{"configurationRank":36,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"muse-spark-1-3","shortName":"Muse Spark 1.3 (max)","terminalBench40Fraction":0.333333333333333},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:458e21cad837a439","board":"aa_terminal_bench_4","model":"glm-5.3-flash","variant":"Artificial Analysis - GLM-5.3-Flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":32.8283,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GLM-5.3-Flash","score":32.8283,"secondary":{"configurationRank":37,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"glm-5-3-flash","shortName":"GLM-5.3-Flash","terminalBench40Fraction":0.328282828282828},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:109f44cf79bff6a4","board":"aa_terminal_bench_4","model":"claude-opus-5.5","variant":"Artificial Analysis - Claude Opus 5.5 (low with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":31.3131,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5.5 (low with fallback)","score":31.3131,"secondary":{"configurationRank":38,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"claude-opus-5-5-low","shortName":"Claude Opus 5.5 (low with fallback)","terminalBench40Fraction":0.313131313131313},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:11dc57e4e600b27d","board":"aa_terminal_bench_4","model":"gpt-6.1-sol","variant":"Artificial Analysis - GPT-6.1 Sol (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":30.8081,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6.1 Sol (low)","score":30.8081,"secondary":{"configurationRank":39,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-6-1-sol-low","shortName":"GPT-6.1 Sol (low)","terminalBench40Fraction":0.308080808080808},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:ba48609dee4ff716","board":"aa_terminal_bench_4","model":"gpt-6-sol","variant":"Artificial Analysis - GPT-6 Sol (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":30.303,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Sol (xhigh)","score":30.303,"secondary":{"configurationRank":40,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-6-sol-xhigh","shortName":"GPT-6 Sol (xhigh)","terminalBench40Fraction":0.303030303030303},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:290fa3a81acd0ea0","board":"aa_terminal_bench_4","model":"claude-sonnet-5.5","variant":"Artificial Analysis - Claude Sonnet 5.5 (medium with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":29.798,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5.5 (medium with fallback)","score":29.798,"secondary":{"configurationRank":41,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"claude-sonnet-5-5-medium","shortName":"Claude Sonnet 5.5 (medium with fallback)","terminalBench40Fraction":0.297979797979798},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:1e5a05b16aa7028a","board":"aa_terminal_bench_4","model":"deepseek-v4.1-flash","variant":"Artificial Analysis - DeepSeek V4.1 Flash (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":26.7677,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4.1-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4.1 Flash (max)","score":26.7677,"secondary":{"configurationRank":42,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"deepseek-v4-1-flash","shortName":"DeepSeek V4.1 Flash (max)","terminalBench40Fraction":0.267676767676768},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a61fb5355f7888c8","board":"aa_terminal_bench_4","model":"claude-opus-5","variant":"Artificial Analysis - Claude Opus 5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":26.2626,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 5 (low)","score":26.2626,"secondary":{"configurationRank":43,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"claude-opus-5-low","shortName":"Claude Opus 5 (low)","terminalBench40Fraction":0.262626262626263},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:2ac75acfdd5556ef","board":"aa_terminal_bench_4","model":"gpt-6-sol","variant":"Artificial Analysis - GPT-6 Sol (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":26.2626,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Sol (high)","score":26.2626,"secondary":{"configurationRank":44,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-6-sol-high","shortName":"GPT-6 Sol (high)","terminalBench40Fraction":0.262626262626263},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:24968b82abf21817","board":"aa_terminal_bench_4","model":"grok-4.7","variant":"Artificial Analysis - Grok 4.7 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":25.7576,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.7","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.7 (xhigh)","score":25.7576,"secondary":{"configurationRank":45,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"grok-4-7","shortName":"Grok 4.7 (xhigh)","terminalBench40Fraction":0.257575757575758},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:131082f722734b10","board":"aa_terminal_bench_4","model":"grok-4.7","variant":"Artificial Analysis - Grok 4.7 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":24.7475,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.7","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.7 (high)","score":24.7475,"secondary":{"configurationRank":47,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"grok-4-7-high","shortName":"Grok 4.7 (high)","terminalBench40Fraction":0.247474747474747},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:f850c291e94d4eee","board":"aa_terminal_bench_4","model":"gpt-5.6-sol","variant":"Artificial Analysis - GPT-5.6 Sol (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":24.7475,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Sol (xhigh)","score":24.7475,"secondary":{"configurationRank":48,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-5-6-sol-xhigh","shortName":"GPT-5.6 Sol (xhigh)","terminalBench40Fraction":0.247474747474747},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:f1477201d34edbb5","board":"aa_terminal_bench_4","model":"mimo-v2.6-flash","variant":"Artificial Analysis - MiMo-V2.6-Flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":22.7273,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.6-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - MiMo-V2.6-Flash","score":22.7273,"secondary":{"configurationRank":49,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"mimo-v2-6-flash","shortName":"MiMo-V2.6-Flash","terminalBench40Fraction":0.227272727272727},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:6569cb5d8342ed58","board":"aa_terminal_bench_4","model":"claude-opus-4.8","variant":"Artificial Analysis - Claude Opus 4.8 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":21.7172,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-4.8","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Opus 4.8 (max)","score":21.7172,"secondary":{"configurationRank":50,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"claude-opus-4-8","shortName":"Claude Opus 4.8 (max)","terminalBench40Fraction":0.217171717171717},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0493139313b0b012","board":"aa_terminal_bench_4","model":"grok-4.6","variant":"Artificial Analysis - Grok 4.6 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":21.2121,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.6 (high)","score":21.2121,"secondary":{"configurationRank":51,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"grok-4-6","shortName":"Grok 4.6 (high)","terminalBench40Fraction":0.212121212121212},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:f0f6b217be9a4963","board":"aa_terminal_bench_4","model":"gpt-5.6-sol","variant":"Artificial Analysis - GPT-5.6 Sol (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":20.7071,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Sol (high)","score":20.7071,"secondary":{"configurationRank":52,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-5-6-sol-high","shortName":"GPT-5.6 Sol (high)","terminalBench40Fraction":0.207070707070707},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:4ae9b6951364c3e4","board":"aa_terminal_bench_4","model":"claude-sonnet-5.5","variant":"Artificial Analysis - Claude Sonnet 5.5 (low with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":20.7071,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5.5 (low with fallback)","score":20.7071,"secondary":{"configurationRank":53,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"claude-sonnet-5-5-low","shortName":"Claude Sonnet 5.5 (low with fallback)","terminalBench40Fraction":0.207070707070707},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a1eae5d167a32a34","board":"aa_terminal_bench_4","model":"gemini-3.8-flash","variant":"Artificial Analysis - Gemini 3.8 Flash (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":19.697,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Gemini 3.8 Flash (medium)","score":19.697,"secondary":{"configurationRank":54,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gemini-3-8-flash-medium","shortName":"Gemini 3.8 Flash (medium)","terminalBench40Fraction":0.196969696969697},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0a0853cb96c8ece1","board":"aa_terminal_bench_4","model":"gemini-3.8-flash","variant":"Artificial Analysis - Gemini 3.8 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":19.697,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Gemini 3.8 Flash (high)","score":19.697,"secondary":{"configurationRank":55,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gemini-3-8-flash","shortName":"Gemini 3.8 Flash (high)","terminalBench40Fraction":0.196969696969697},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0487a0a408dccb82","board":"aa_terminal_bench_4","model":"qwen3.8-max","variant":"Artificial Analysis - Qwen3.8 Max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":18.6869,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-max","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.8 Max","score":18.6869,"secondary":{"configurationRank":56,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"qwen3-8-max-0803","shortName":"Qwen3.8 Max","terminalBench40Fraction":0.186868686868687},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:5b54a5c8b54994f4","board":"aa_terminal_bench_4","model":"gpt-6-sol","variant":"Artificial Analysis - GPT-6 Sol (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":18.6869,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Sol (medium)","score":18.6869,"secondary":{"configurationRank":57,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-6-sol-medium","shortName":"GPT-6 Sol (medium)","terminalBench40Fraction":0.186868686868687},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:27d67549b5ee1d8d","board":"aa_terminal_bench_4","model":"grok-4.6","variant":"Artificial Analysis - Grok 4.6 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":17.1717,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.6 (xhigh)","score":17.1717,"secondary":{"configurationRank":58,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"grok-4-6-xhigh","shortName":"Grok 4.6 (xhigh)","terminalBench40Fraction":0.171717171717172},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:77d9a42a8772a37d","board":"aa_terminal_bench_4","model":"muse-spark-1.3","variant":"Artificial Analysis - Muse Spark 1.3 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":16.6667,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Muse Spark 1.3 (xhigh)","score":16.6667,"secondary":{"configurationRank":59,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"muse-spark-1-3-xhigh","shortName":"Muse Spark 1.3 (xhigh)","terminalBench40Fraction":0.166666666666667},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:b426d7104b617616","board":"aa_terminal_bench_4","model":"gpt-5.5","variant":"Artificial Analysis - GPT-5.5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":14.6465,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.5 (xhigh)","score":14.6465,"secondary":{"configurationRank":60,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-5-5","shortName":"GPT-5.5 (xhigh)","terminalBench40Fraction":0.146464646464646},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:de8c4cd8b48f6964","board":"aa_terminal_bench_4","model":"gpt-5.6-sol","variant":"Artificial Analysis - GPT-5.6 Sol (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":14.6465,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Sol (medium)","score":14.6465,"secondary":{"configurationRank":61,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-5-6-sol-medium","shortName":"GPT-5.6 Sol (medium)","terminalBench40Fraction":0.146464646464646},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:c2972229b7214dfc","board":"aa_terminal_bench_4","model":"deepseek-v4-pro","variant":"Artificial Analysis - DeepSeek V4 Pro (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":14.6465,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Pro (max)","score":14.6465,"secondary":{"configurationRank":62,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"deepseek-v4-pro-0424","shortName":"DeepSeek V4 Pro (max)","terminalBench40Fraction":0.146464646464646},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:8a2bf906a5cca32e","board":"aa_terminal_bench_4","model":"claude-sonnet-5","variant":"Artificial Analysis - Claude Sonnet 5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":14.1414,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5 (max)","score":14.1414,"secondary":{"configurationRank":63,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"claude-sonnet-5","shortName":"Claude Sonnet 5 (max)","terminalBench40Fraction":0.141414141414141},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:92c8207e8ca30249","board":"aa_terminal_bench_4","model":"deepseek-v4-pro","variant":"Artificial Analysis - DeepSeek V4 Pro 0813 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":14.1414,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Pro 0813 (max)","score":14.1414,"secondary":{"configurationRank":64,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"deepseek-v4-pro","shortName":"DeepSeek V4 Pro 0813 (max)","terminalBench40Fraction":0.141414141414141},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:69dd3b76427af507","board":"aa_terminal_bench_4","model":"gemini-3.7-flash","variant":"Artificial Analysis - Gemini 3.7 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":13.6364,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.7-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Gemini 3.7 Flash (high)","score":13.6364,"secondary":{"configurationRank":65,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gemini-3-7-flash","shortName":"Gemini 3.7 Flash (high)","terminalBench40Fraction":0.136363636363636},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:008592dee2abfbd4","board":"aa_terminal_bench_4","model":"grok-4.6","variant":"Artificial Analysis - Grok 4.6 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":13.1313,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.6 (medium)","score":13.1313,"secondary":{"configurationRank":66,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"grok-4-6-medium","shortName":"Grok 4.6 (medium)","terminalBench40Fraction":0.131313131313131},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:ba96dfd07fb6e2bc","board":"aa_terminal_bench_4","model":"gpt-6-sol","variant":"Artificial Analysis - GPT-6 Sol (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":13.1313,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Sol (non-reasoning)","score":13.1313,"secondary":{"configurationRank":67,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"gpt-6-sol-non-reasoning","shortName":"GPT-6 Sol (non-reasoning)","terminalBench40Fraction":0.131313131313131},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:cc7c150caad9b305","board":"aa_terminal_bench_4","model":"kimi-k3","variant":"Artificial Analysis - Kimi K3 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":12.6263,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Kimi K3 (low)","score":12.6263,"secondary":{"configurationRank":68,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"kimi-k3-low","shortName":"Kimi K3 (low)","terminalBench40Fraction":0.126262626262626},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:4d4153eb38f2e84f","board":"aa_terminal_bench_4","model":"gpt-6-luna","variant":"Artificial Analysis - GPT-6 Luna (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":12.6263,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Luna (max)","score":12.6263,"secondary":{"configurationRank":69,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-6-luna","shortName":"GPT-6 Luna (max)","terminalBench40Fraction":0.126262626262626},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a6ac642cac2ac396","board":"aa_terminal_bench_4","model":"kimi-k3","variant":"Artificial Analysis - Kimi K3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":12.6263,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Kimi K3 (max)","score":12.6263,"secondary":{"configurationRank":71,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"kimi-k3","shortName":"Kimi K3 (max)","terminalBench40Fraction":0.126262626262626},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:2bd2f5c3a28cefaa","board":"aa_terminal_bench_4","model":"deepseek-v4-flash-vision-exp","variant":"Artificial Analysis - DeepSeek V4 Flash Vision (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":12.1212,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash-vision-exp","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Flash Vision (max)","score":12.1212,"secondary":{"configurationRank":72,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"deepseek-v4-flash-vision","shortName":"DeepSeek V4 Flash Vision (max)","terminalBench40Fraction":0.121212121212121},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a807bd1873a40322","board":"aa_terminal_bench_4","model":"deepseek-v4-flash","variant":"Artificial Analysis - DeepSeek V4 Flash 0731 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":12.1212,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Flash 0731 (max)","score":12.1212,"secondary":{"configurationRank":73,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"deepseek-v4-flash","shortName":"DeepSeek V4 Flash 0731 (max)","terminalBench40Fraction":0.121212121212121},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:9c11404866580c5d","board":"aa_terminal_bench_4","model":"gpt-5.6-luna","variant":"Artificial Analysis - GPT-5.6 Luna (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":11.6162,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Luna (max)","score":11.6162,"secondary":{"configurationRank":74,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"gpt-5-6-luna","shortName":"GPT-5.6 Luna (max)","terminalBench40Fraction":0.116161616161616},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:68bfb23c4047a7a3","board":"aa_terminal_bench_4","model":"grok-4.5","variant":"Artificial Analysis - Grok 4.5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":10.6061,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.5 (high)","score":10.6061,"secondary":{"configurationRank":76,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"grok-4-5","shortName":"Grok 4.5 (high)","terminalBench40Fraction":0.106060606060606},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:653b68c5fe321a32","board":"aa_terminal_bench_4","model":"gpt-5.6-terra","variant":"Artificial Analysis - GPT-5.6 Terra (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":10.101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Terra (xhigh)","score":10.101,"secondary":{"configurationRank":77,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-5-6-terra-xhigh","shortName":"GPT-5.6 Terra (xhigh)","terminalBench40Fraction":0.101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:76b42aa2f87efd5c","board":"aa_terminal_bench_4","model":"gemini-3.8-flash","variant":"Artificial Analysis - Gemini 3.8 Flash (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":10.101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Gemini 3.8 Flash (low)","score":10.101,"secondary":{"configurationRank":78,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gemini-3-8-flash-low","shortName":"Gemini 3.8 Flash (low)","terminalBench40Fraction":0.101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:af9e50a31779df75","board":"aa_terminal_bench_4","model":"gpt-6-sol","variant":"Artificial Analysis - GPT-6 Sol (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":9.0909,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Sol (low)","score":9.0909,"secondary":{"configurationRank":79,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-6-sol-low","shortName":"GPT-6 Sol (low)","terminalBench40Fraction":0.0909090909090909},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:7bff670f3c107bb6","board":"aa_terminal_bench_4","model":"gpt-5.5","variant":"Artificial Analysis - GPT-5.5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":9.0909,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.5 (high)","score":9.0909,"secondary":{"configurationRank":80,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-5-5-high","shortName":"GPT-5.5 (high)","terminalBench40Fraction":0.0909090909090909},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:3956f4765e345bfa","board":"aa_terminal_bench_4","model":"gpt-6-luna","variant":"Artificial Analysis - GPT-6 Luna (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":8.0808,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Luna (xhigh)","score":8.0808,"secondary":{"configurationRank":81,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-6-luna-xhigh","shortName":"GPT-6 Luna (xhigh)","terminalBench40Fraction":0.0808080808080808},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:af37ce7e1e85aaba","board":"aa_terminal_bench_4","model":"muse-spark-1.2","variant":"Artificial Analysis - Muse Spark 1.2 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":7.0707,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.2","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Muse Spark 1.2 (xhigh)","score":7.0707,"secondary":{"configurationRank":82,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"muse-spark-1-2","shortName":"Muse Spark 1.2 (xhigh)","terminalBench40Fraction":0.0707070707070707},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:ec826f6447a6ed8b","board":"aa_terminal_bench_4","model":"gemini-3.6-flash","variant":"Artificial Analysis - Gemini 3.6 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":7.0707,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.6-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Gemini 3.6 Flash (high)","score":7.0707,"secondary":{"configurationRank":83,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gemini-3-6-flash","shortName":"Gemini 3.6 Flash (high)","terminalBench40Fraction":0.0707070707070707},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:58f8a2ad361cbfac","board":"aa_terminal_bench_4","model":"claude-sonnet-5","variant":"Artificial Analysis - Claude Sonnet 5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":7.0707,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5 (xhigh)","score":7.0707,"secondary":{"configurationRank":84,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"claude-sonnet-5-xhigh","shortName":"Claude Sonnet 5 (xhigh)","terminalBench40Fraction":0.0707070707070707},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:5b3cd3f66e09bf91","board":"aa_terminal_bench_4","model":"deepseek-v4.1-flash","variant":"Artificial Analysis - DeepSeek V4.1 Flash (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":5.5556,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4.1-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4.1 Flash (non-reasoning)","score":5.5556,"secondary":{"configurationRank":87,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"deepseek-v4-1-flash-non-reasoning","shortName":"DeepSeek V4.1 Flash (non-reasoning)","terminalBench40Fraction":0.0555555555555556},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:c563447fe73f34f6","board":"aa_terminal_bench_4","model":"qwen3.8-27b","variant":"Artificial Analysis - Qwen3.8 27B (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":5.5556,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.8 27B (xhigh)","score":5.5556,"secondary":{"configurationRank":88,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"qwen3-8-27b","shortName":"Qwen3.8 27B (xhigh)","terminalBench40Fraction":0.0555555555555556},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:d9a51d09614610e8","board":"aa_terminal_bench_4","model":"qwen3.8-27b","variant":"Artificial Analysis - Qwen3.8 27B (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":5.0505,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.8 27B (medium)","score":5.0505,"secondary":{"configurationRank":89,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"qwen3-8-27b-medium","shortName":"Qwen3.8 27B (medium)","terminalBench40Fraction":0.0505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:52b22206889b7f5e","board":"aa_terminal_bench_4","model":"gpt-5.5","variant":"Artificial Analysis - GPT-5.5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":5.0505,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.5 (medium)","score":5.0505,"secondary":{"configurationRank":90,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-5-5-medium","shortName":"GPT-5.5 (medium)","terminalBench40Fraction":0.0505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:4583c02c014dc3d1","board":"aa_terminal_bench_4","model":"claude-sonnet-5","variant":"Artificial Analysis - Claude Sonnet 5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":5.0505,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5 (high)","score":5.0505,"secondary":{"configurationRank":91,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"claude-sonnet-5-high","shortName":"Claude Sonnet 5 (high)","terminalBench40Fraction":0.0505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:964384bb8d3b9a13","board":"aa_terminal_bench_4","model":"gpt-6-luna","variant":"Artificial Analysis - GPT-6 Luna (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":4.5455,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Luna (high)","score":4.5455,"secondary":{"configurationRank":92,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-6-luna-high","shortName":"GPT-6 Luna (high)","terminalBench40Fraction":0.0454545454545455},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:28b1f129442e7f1a","board":"aa_terminal_bench_4","model":"gemini-3.1-pro","variant":"Artificial Analysis - Gemini 3.1 Pro Preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":4.0404,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.1-pro","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Gemini 3.1 Pro Preview","score":4.0404,"secondary":{"configurationRank":93,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"gemini-3-1-pro-preview","shortName":"Gemini 3.1 Pro Preview","terminalBench40Fraction":0.0404040404040404},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:79ab9eec86d38bba","board":"aa_terminal_bench_4","model":"gpt-5.6-luna","variant":"Artificial Analysis - GPT-5.6 Luna (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":3.5354,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Luna (xhigh)","score":3.5354,"secondary":{"configurationRank":94,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"gpt-5-6-luna-xhigh","shortName":"GPT-5.6 Luna (xhigh)","terminalBench40Fraction":0.0353535353535354},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a2b58faa52397cc0","board":"aa_terminal_bench_4","model":"deepseek-v4-flash","variant":"Artificial Analysis - DeepSeek V4 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":3.0303,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Flash (high)","score":3.0303,"secondary":{"configurationRank":96,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"deepseek-v4-flash-0420-high","shortName":"DeepSeek V4 Flash (high)","terminalBench40Fraction":0.0303030303030303},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:ab17cc8e5487de99","board":"aa_terminal_bench_4","model":"grok-4.6","variant":"Artificial Analysis - Grok 4.6 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":3.0303,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Grok 4.6 (low)","score":3.0303,"secondary":{"configurationRank":97,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"grok-4-6-low","shortName":"Grok 4.6 (low)","terminalBench40Fraction":0.0303030303030303},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0a46e136780aeec1","board":"aa_terminal_bench_4","model":"claude-sonnet-5","variant":"Artificial Analysis - Claude Sonnet 5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":2.5253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5 (low)","score":2.5253,"secondary":{"configurationRank":99,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"claude-sonnet-5-low","shortName":"Claude Sonnet 5 (low)","terminalBench40Fraction":0.0252525252525253},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:62b32b0accfdc3e1","board":"aa_terminal_bench_4","model":"gpt-6-luna","variant":"Artificial Analysis - GPT-6 Luna (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":2.5253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Luna (medium)","score":2.5253,"secondary":{"configurationRank":100,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-6-luna-medium","shortName":"GPT-6 Luna (medium)","terminalBench40Fraction":0.0252525252525253},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a025952071adf945","board":"aa_terminal_bench_4","model":"qwen3.8-27b","variant":"Artificial Analysis - Qwen3.8 27B (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":2.5253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.8 27B (low)","score":2.5253,"secondary":{"configurationRank":101,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"qwen3-8-27b-low","shortName":"Qwen3.8 27B (low)","terminalBench40Fraction":0.0252525252525253},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:f7293c3aff3fb8b9","board":"aa_terminal_bench_4","model":"gpt-5.6-luna","variant":"Artificial Analysis - GPT-5.6 Luna (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":2.5253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Luna (high)","score":2.5253,"secondary":{"configurationRank":102,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-5-6-luna-high","shortName":"GPT-5.6 Luna (high)","terminalBench40Fraction":0.0252525252525253},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:fee6e8fd477b7ac8","board":"aa_terminal_bench_4","model":"deepseek-v4-flash","variant":"Artificial Analysis - DeepSeek V4 Flash (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":2.5253,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Flash (max)","score":2.5253,"secondary":{"configurationRank":103,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"deepseek-v4-flash-0420","shortName":"DeepSeek V4 Flash (max)","terminalBench40Fraction":0.0252525252525253},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:4f6bf2d4de391fa8","board":"aa_terminal_bench_4","model":"minimax-m3","variant":"Artificial Analysis - MiniMax-M3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":2.0202,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"minimax-m3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - MiniMax-M3","score":2.0202,"secondary":{"configurationRank":104,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"minimax-m3","shortName":"MiniMax-M3","terminalBench40Fraction":0.0202020202020202},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:39ff8a3fc7a8b6c2","board":"aa_terminal_bench_4","model":"glm-5.1","variant":"Artificial Analysis - GLM-5.1","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":2.0202,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.1","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GLM-5.1","score":2.0202,"secondary":{"configurationRank":105,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"glm-5-1","shortName":"GLM-5.1","terminalBench40Fraction":0.0202020202020202},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:eecd341c62332c35","board":"aa_terminal_bench_4","model":"claude-sonnet-5","variant":"Artificial Analysis - Claude Sonnet 5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":2.0202,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Claude Sonnet 5 (medium)","score":2.0202,"secondary":{"configurationRank":107,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"claude-sonnet-5-medium","shortName":"Claude Sonnet 5 (medium)","terminalBench40Fraction":0.0202020202020202},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:c911fafd76fe1373","board":"aa_terminal_bench_4","model":"gpt-6-luna","variant":"Artificial Analysis - GPT-6 Luna (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":1.5152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Luna (non-reasoning)","score":1.5152,"secondary":{"configurationRank":108,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"gpt-6-luna-non-reasoning","shortName":"GPT-6 Luna (non-reasoning)","terminalBench40Fraction":0.0151515151515152},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a3c02275461ec0cb","board":"aa_terminal_bench_4","model":"qwen3.7-max","variant":"Artificial Analysis - Qwen3.7 Max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":1.5152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.7-max","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.7 Max","score":1.5152,"secondary":{"configurationRank":109,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"qwen3-7-max","shortName":"Qwen3.7 Max","terminalBench40Fraction":0.0151515151515152},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:8239d3088594c526","board":"aa_terminal_bench_4","model":"gpt-5.6-terra","variant":"Artificial Analysis - GPT-5.6 Terra (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"high","service_mode":null,"score":1.5152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Terra (high)","score":1.5152,"secondary":{"configurationRank":110,"agentHarness":"Artificial Analysis","reasoningEffort":"high","slug":"gpt-5-6-terra-high","shortName":"GPT-5.6 Terra (high)","terminalBench40Fraction":0.0151515151515152},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:b68acd500c6b4161","board":"aa_terminal_bench_4","model":"gpt-5.6-terra","variant":"Artificial Analysis - GPT-5.6 Terra (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":1.5152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Terra (low)","score":1.5152,"secondary":{"configurationRank":112,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-5-6-terra-low","shortName":"GPT-5.6 Terra (low)","terminalBench40Fraction":0.0151515151515152},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:79eac9decde86827","board":"aa_terminal_bench_4","model":"deepseek-v4-pro","variant":"Artificial Analysis - DeepSeek V4 Pro 0813 (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":1.5152,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - DeepSeek V4 Pro 0813 (non-reasoning)","score":1.5152,"secondary":{"configurationRank":113,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"deepseek-v4-pro-non-reasoning","shortName":"DeepSeek V4 Pro 0813 (non-reasoning)","terminalBench40Fraction":0.0151515151515152},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a5ddeab611927928","board":"aa_terminal_bench_4","model":"gpt-5.6-sol","variant":"Artificial Analysis - GPT-5.6 Sol (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Sol (low)","score":1.0101,"secondary":{"configurationRank":114,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-5-6-sol-low","shortName":"GPT-5.6 Sol (low)","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:8b8fcc0fa14e07eb","board":"aa_terminal_bench_4","model":"inkling","variant":"Artificial Analysis - Inkling (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"xhigh","service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"inkling","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Inkling (xhigh)","score":1.0101,"secondary":{"configurationRank":116,"agentHarness":"Artificial Analysis","reasoningEffort":"xhigh","slug":"inkling","shortName":"Inkling (xhigh)","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:9dda7394ba62eb25","board":"aa_terminal_bench_4","model":"gpt-5.6-terra","variant":"Artificial Analysis - GPT-5.6 Terra (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Terra (medium)","score":1.0101,"secondary":{"configurationRank":118,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-5-6-terra-medium","shortName":"GPT-5.6 Terra (medium)","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:d0b5249bef72ad97","board":"aa_terminal_bench_4","model":"inkling-small","variant":"Artificial Analysis - Inkling Small","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"inkling-small","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Inkling Small","score":1.0101,"secondary":{"configurationRank":120,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"inkling-small","shortName":"Inkling Small","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:c9067f00834590ba","board":"aa_terminal_bench_4","model":"kimi-k2.7-code","variant":"Artificial Analysis - Kimi K2.7 Code","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.7-code","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Kimi K2.7 Code","score":1.0101,"secondary":{"configurationRank":121,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"kimi-k2-7-code","shortName":"Kimi K2.7 Code","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:3d9d54fa83beecde","board":"aa_terminal_bench_4","model":"qwen3.7-plus","variant":"Artificial Analysis - Qwen3.7 Plus","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.7-plus","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.7 Plus","score":1.0101,"secondary":{"configurationRank":122,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"qwen3-7-plus","shortName":"Qwen3.7 Plus","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:a734c5ed959dfebb","board":"aa_terminal_bench_4","model":"gpt-5.6-luna","variant":"Artificial Analysis - GPT-5.6 Luna (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Luna (non-reasoning)","score":1.0101,"secondary":{"configurationRank":123,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"gpt-5-6-luna-non-reasoning","shortName":"GPT-5.6 Luna (non-reasoning)","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:9ad8c75e6b68658f","board":"aa_terminal_bench_4","model":"glm-5.2","variant":"Artificial Analysis - GLM-5.2 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"max","service_mode":null,"score":1.0101,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.2","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GLM-5.2 (max)","score":1.0101,"secondary":{"configurationRank":124,"agentHarness":"Artificial Analysis","reasoningEffort":"max","slug":"glm-5-2","shortName":"GLM-5.2 (max)","terminalBench40Fraction":0.0101010101010101},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:ca3671e92d985c3e","board":"aa_terminal_bench_4","model":"kimi-k2.6","variant":"Artificial Analysis - Kimi K2.6","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0.5051,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k2.6","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Kimi K2.6","score":0.5051,"secondary":{"configurationRank":126,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"kimi-k2-6","shortName":"Kimi K2.6","terminalBench40Fraction":0.00505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:1837d1adb8970232","board":"aa_terminal_bench_4","model":"gpt-5.6-luna","variant":"Artificial Analysis - GPT-5.6 Luna (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":0.5051,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Luna (medium)","score":0.5051,"secondary":{"configurationRank":133,"agentHarness":"Artificial Analysis","reasoningEffort":"medium","slug":"gpt-5-6-luna-medium","shortName":"GPT-5.6 Luna (medium)","terminalBench40Fraction":0.00505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:3d0d41d5c6a6aa9b","board":"aa_terminal_bench_4","model":"nemotron-3-ultra","variant":"Artificial Analysis - Nemotron 3 Ultra","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0.5051,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"nemotron-3-ultra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Nemotron 3 Ultra","score":0.5051,"secondary":{"configurationRank":135,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"nvidia-nemotron-3-ultra-550b-a55b","shortName":"Nemotron 3 Ultra","terminalBench40Fraction":0.00505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:2aa44c238da36b9a","board":"aa_terminal_bench_4","model":"gpt-5.6-terra","variant":"Artificial Analysis - GPT-5.6 Terra (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0.5051,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-terra","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Terra (non-reasoning)","score":0.5051,"secondary":{"configurationRank":136,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"gpt-5-6-terra-non-reasoning","shortName":"GPT-5.6 Terra (non-reasoning)","terminalBench40Fraction":0.00505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:7e5e9735030b01c8","board":"aa_terminal_bench_4","model":"hy3","variant":"Artificial Analysis - Hy3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0.5051,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"hy3","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Hy3","score":0.5051,"secondary":{"configurationRank":137,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"hy3","shortName":"Hy3","terminalBench40Fraction":0.00505050505050505},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:bfe134e562d0cd53","board":"aa_terminal_bench_4","model":"mimo-v2.5-pro","variant":"Artificial Analysis - MiMo-V2.5-Pro","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.5-pro","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - MiMo-V2.5-Pro","score":0,"secondary":{"configurationRank":141,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"mimo-v2-5-pro","shortName":"MiMo-V2.5-Pro","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:2f7e69749a01454b","board":"aa_terminal_bench_4","model":"gpt-5.6-luna","variant":"Artificial Analysis - GPT-5.6 Luna (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-5.6 Luna (low)","score":0,"secondary":{"configurationRank":143,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-5-6-luna-low","shortName":"GPT-5.6 Luna (low)","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:5a02fb72c74be509","board":"aa_terminal_bench_4","model":"mimo-v2.5","variant":"Artificial Analysis - MiMo-V2.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mimo-v2.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - MiMo-V2.5","score":0,"secondary":{"configurationRank":153,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"mimo-v2-5-0424","shortName":"MiMo-V2.5","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:7f4fd5ee28b042c6","board":"aa_terminal_bench_4","model":"qwen3.8-27b","variant":"Artificial Analysis - Qwen3.8 27B (non-reasoning)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-27b","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Qwen3.8 27B (non-reasoning)","score":0,"secondary":{"configurationRank":161,"agentHarness":"Artificial Analysis","reasoningEffort":"non-reasoning","slug":"qwen3-8-27b-non-reasoning","shortName":"Qwen3.8 27B (non-reasoning)","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:0ba9180d8eb1c0d7","board":"aa_terminal_bench_4","model":"minimax-m2.7","variant":"Artificial Analysis - MiniMax-M2.7","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"minimax-m2.7","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - MiniMax-M2.7","score":0,"secondary":{"configurationRank":169,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"minimax-m2-7","shortName":"MiniMax-M2.7","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:d0a6a751fe4ac57c","board":"aa_terminal_bench_4","model":"longcat-2.0","variant":"Artificial Analysis - LongCat 2.0","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":null,"service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"longcat-2.0","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - LongCat 2.0","score":0,"secondary":{"configurationRank":171,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"longcat-2-0","shortName":"LongCat 2.0","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:1f723b35d187e5ec","board":"aa_terminal_bench_4","model":"gpt-6-luna","variant":"Artificial Analysis - GPT-6 Luna (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"low","service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - GPT-6 Luna (low)","score":0,"secondary":{"configurationRank":198,"agentHarness":"Artificial Analysis","reasoningEffort":"low","slug":"gpt-6-luna-low","shortName":"GPT-6 Luna (low)","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_terminal_bench_4:d8495a02281ad221","board":"aa_terminal_bench_4","model":"mistral-medium-3.5","variant":"Artificial Analysis - Mistral Medium 3.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Artificial Analysis","reasoning_effort":"medium","service_mode":null,"score":0,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/leaderboards/models","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"mistral-medium-3.5","boardId":"aa_terminal_bench_4","variantLabel":"Artificial Analysis - Mistral Medium 3.5","score":0,"secondary":{"configurationRank":202,"agentHarness":"Artificial Analysis","reasoningEffort":null,"slug":"mistral-medium-3-5","shortName":"Mistral Medium 3.5","terminalBench40Fraction":0},"source":"https://artificialanalysis.ai/leaderboards/models","checkedAt":"2026-10-01","note":"AA 自家 harness 独立实测的 Terminal-Bench 4.0（models 榜载荷 terminalBench40 字段，fraction×100）；与 tbench.ai 官方榜分开计分。"}},{"configuration_id":"aa_coding_agent_index:608ea08d36d3bc63","board":"aa_coding_agent_index","model":"claude-sonnet-5.5","variant":"Claude Code - Sonnet 5.5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":68.3578,"score_low":null,"score_high":null,"mean_cost_usd_per_task":14.1908,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Sonnet 5.5 (max)","score":68.3578,"secondary":{"configurationRank":1,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"private-anthropic_claude-beignet-eap-max","displayModel":"Sonnet 5.5 (max)","isUnavailable":false,"variantOf":null,"deepSWE":71.9764,"terminalBenchV4":66.1616,"sweAtlasQnA":66.9355,"meanCostUsdPerTask":14.1908,"agentWallTimeSec":5244.9654,"meanSteps":265.9307,"meanTotalTokens":27742337.2338,"cacheHitRate":0.9506,"hostConfigId":"a2c87c062f3cef73d8525e7578f14742","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:d3474ba216efd88d","board":"aa_coding_agent_index","model":"claude-opus-5.5","variant":"Claude Code - Opus 5.5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":65.9886,"score_low":null,"score_high":null,"mean_cost_usd_per_task":13.0364,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5.5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Opus 5.5 (max)","score":65.9886,"secondary":{"configurationRank":2,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"anthropic_claude-opus-5-5","displayModel":"Opus 5.5 (max)","isUnavailable":false,"variantOf":null,"deepSWE":68.4366,"terminalBenchV4":63.1313,"sweAtlasQnA":66.3978,"meanCostUsdPerTask":13.0364,"agentWallTimeSec":3867.0923,"meanSteps":155.0957,"meanTotalTokens":15551383.22,"cacheHitRate":0.95,"hostConfigId":"ab5074e404be48bbd4ac3c488ad9a3e2","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:9f60c3eaad821549","board":"aa_coding_agent_index","model":"gpt-6.1-sol","variant":"Codex - GPT-6.1 Sol (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"xhigh","service_mode":null,"score":62.9078,"score_low":null,"score_high":null,"mean_cost_usd_per_task":1.0397,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6.1 Sol (xhigh) ({'reasoning_effort': 'xhigh'})","score":62.9078,"secondary":{"configurationRank":4,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_quasar-alpha","displayModel":"GPT-6.1 Sol (xhigh)","isUnavailable":false,"variantOf":null,"deepSWE":73.1563,"terminalBenchV4":54.5455,"sweAtlasQnA":61.0215,"meanCostUsdPerTask":1.0397,"agentWallTimeSec":931.3823,"meanSteps":40.055,"meanTotalTokens":3212528.1095,"cacheHitRate":0.9328,"hostConfigId":"ed8c1d0b0a2fe4560bbfa0a9f47e8941","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:54bdbbcca087e6ff","board":"aa_coding_agent_index","model":"claude-sonnet-5.5","variant":"Claude Code - Sonnet 5.5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"xhigh","service_mode":null,"score":62.8714,"score_low":null,"score_high":null,"mean_cost_usd_per_task":3.3334,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Sonnet 5.5 (xhigh)","score":62.8714,"secondary":{"configurationRank":5,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"private-anthropic_claude-beignet-eap-xhigh","displayModel":"Sonnet 5.5 (xhigh)","isUnavailable":false,"variantOf":null,"deepSWE":68.4366,"terminalBenchV4":58.0808,"sweAtlasQnA":62.0968,"meanCostUsdPerTask":3.3334,"agentWallTimeSec":1619.0135,"meanSteps":78.3168,"meanTotalTokens":7087678.396,"cacheHitRate":0.9575,"hostConfigId":"666605c66ad569c503ce53c01682e285","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:f882189259104e87","board":"aa_coding_agent_index","model":"claude-fable-5.1","variant":"Claude Code - Fable 5.1 (max) (with fallback)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":62.2225,"score_low":null,"score_high":null,"mean_cost_usd_per_task":12.3881,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Fable 5.1 (max) (with fallback)","score":62.2225,"secondary":{"configurationRank":6,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"anthropic_claude-fable-5-1","displayModel":"Fable 5.1 (max) (with fallback)","isUnavailable":false,"variantOf":null,"deepSWE":64.3068,"terminalBenchV4":57.5758,"sweAtlasQnA":64.7849,"meanCostUsdPerTask":12.3881,"agentWallTimeSec":2089.9001,"meanSteps":36.6568,"meanTotalTokens":5715253.8592,"cacheHitRate":0.916,"hostConfigId":"29a1cdad0d140781d05b9b4746cfdd0d","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.263","dateReleased":"2026-09-06"},"max":{"version":"2.1.263","dateReleased":"2026-09-06"}},"swe-atlas-qna":{"min":{"version":"2.1.259","dateReleased":"2026-09-02"},"max":{"version":"2.1.259","dateReleased":"2026-09-02"}},"terminal-bench-v4":{"min":{"version":"2.1.263","dateReleased":"2026-09-06"},"max":{"version":"2.1.263","dateReleased":"2026-09-06"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:1041b2cdd9886064","board":"aa_coding_agent_index","model":"claude-fable-5.1","variant":"Devin Fusion CLI - Claude Fable 5.1 XHigh + SWE-2 Medium","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Devin Fusion CLI","reasoning_effort":"xhigh","service_mode":null,"score":61.6826,"score_low":null,"score_high":null,"mean_cost_usd_per_task":7.8971,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-fable-5.1","boardId":"aa_coding_agent_index","variantLabel":"Devin Fusion CLI - Claude Fable 5.1 XHigh + SWE-2 Medium","score":61.6826,"secondary":{"configurationRank":7,"agentHarness":"Devin Fusion CLI","provider":"cognition","hostModelSlug":"cognition_fusion-claude-fable-5-1-xhigh-sidekick-penguin-medium","displayModel":"Claude Fable 5.1 XHigh + SWE-2 Medium","isUnavailable":false,"variantOf":null,"deepSWE":63.1268,"terminalBenchV4":56.0606,"sweAtlasQnA":65.8602,"meanCostUsdPerTask":7.8971,"agentWallTimeSec":2146.9921,"meanSteps":99.5314,"meanTotalTokens":9691548.1738,"cacheHitRate":0.9585,"hostConfigId":"34970c963bdfd318e46205514d46ea4f","versions":{"deep-swe-v1.1":{"min":{"version":"3000.10.7","dateReleased":null},"max":{"version":"3000.10.7","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"3000.10.7","dateReleased":null},"max":{"version":"3000.10.7","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"3000.10.7","dateReleased":null},"max":{"version":"3000.10.7","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:6ee94ff7d755eeee","board":"aa_coding_agent_index","model":"gpt-6-astra","variant":"Codex - GPT-6 Astra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":61.645,"score_low":null,"score_high":null,"mean_cost_usd_per_task":7.4671,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6 Astra (max)","score":61.645,"secondary":{"configurationRank":8,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_vega-alpha","displayModel":"GPT-6 Astra (max)","isUnavailable":false,"variantOf":null,"deepSWE":67.5516,"terminalBenchV4":55.5556,"sweAtlasQnA":61.828,"meanCostUsdPerTask":7.4671,"agentWallTimeSec":1762.033,"meanSteps":39.3784,"meanTotalTokens":3346611.0122,"cacheHitRate":0.9336,"hostConfigId":"9e810ff3c6cce5ebaaff7c744c687a47","versions":{"deep-swe-v1.1":{"min":{"version":"0.153.4","dateReleased":"2026-09-04"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}},"swe-atlas-qna":{"min":{"version":"0.151.0","dateReleased":"2026-08-29"},"max":{"version":"0.151.0","dateReleased":"2026-08-29"}},"terminal-bench-v4":{"min":{"version":"0.153.4","dateReleased":"2026-09-04"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:0e50f95e98c947c2","board":"aa_coding_agent_index","model":"gpt-6.1-sol","variant":"Codex - GPT-6.1 Sol (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"medium","service_mode":null,"score":61.4147,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.7047,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6.1 Sol (medium) ({'reasoning_effort': 'medium'})","score":61.4147,"secondary":{"configurationRank":9,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_quasar-alpha","displayModel":"GPT-6.1 Sol (medium)","isUnavailable":false,"variantOf":null,"deepSWE":71.9764,"terminalBenchV4":51.5152,"sweAtlasQnA":60.7527,"meanCostUsdPerTask":0.7047,"agentWallTimeSec":651.3229,"meanSteps":37.989,"meanTotalTokens":2298278.7454,"cacheHitRate":0.924,"hostConfigId":"93acd59f31f8e5478686d650aa650f79","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:6adf1a461abaf0f2","board":"aa_coding_agent_index","model":"gpt-6.1-sol","variant":"Codex - GPT-6.1 Sol (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"high","service_mode":null,"score":60.1492,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.8892,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6.1 Sol (high) ({'reasoning_effort': 'high'})","score":60.1492,"secondary":{"configurationRank":10,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_quasar-alpha","displayModel":"GPT-6.1 Sol (high)","isUnavailable":false,"variantOf":null,"deepSWE":70.5015,"terminalBenchV4":50,"sweAtlasQnA":59.9462,"meanCostUsdPerTask":0.8892,"agentWallTimeSec":799.7652,"meanSteps":38.9791,"meanTotalTokens":2819911.9904,"cacheHitRate":0.9278,"hostConfigId":"4aee8aa2d4a0bad6068644f40edae235","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:718b49f369ed6f33","board":"aa_coding_agent_index","model":"gpt-6.1-sol","variant":"Codex - GPT-6.1 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":60.1475,"score_low":null,"score_high":null,"mean_cost_usd_per_task":1.5541,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6.1 Sol (max)","score":60.1475,"secondary":{"configurationRank":11,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_quasar-alpha","displayModel":"GPT-6.1 Sol (max)","isUnavailable":false,"variantOf":null,"deepSWE":69.6165,"terminalBenchV4":53.0303,"sweAtlasQnA":57.7957,"meanCostUsdPerTask":1.5541,"agentWallTimeSec":1463.0464,"meanSteps":40.4565,"meanTotalTokens":3999501.6108,"cacheHitRate":0.9282,"hostConfigId":"ea296ce785422d3b97cfff4427f2e0d2","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:e0b71dfaa4f1714e","board":"aa_coding_agent_index","model":"claude-opus-5","variant":"Claude Code - Opus 5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":59.7264,"score_low":null,"score_high":null,"mean_cost_usd_per_task":10.7892,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-opus-5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Opus 5 (max)","score":59.7264,"secondary":{"configurationRank":12,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"anthropic_claude-opus-5","displayModel":"Opus 5 (max)","isUnavailable":false,"variantOf":null,"deepSWE":62.5369,"terminalBenchV4":54.5455,"sweAtlasQnA":62.0968,"meanCostUsdPerTask":10.7892,"agentWallTimeSec":2515.5402,"meanSteps":152.1304,"meanTotalTokens":11374850.489,"cacheHitRate":0.9698,"hostConfigId":"409518818f1e9ad21995d70631ab05b6","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.263","dateReleased":"2026-09-06"},"max":{"version":"2.1.263","dateReleased":"2026-09-06"}},"swe-atlas-qna":{"min":{"version":"2.1.218","dateReleased":"2026-07-22"},"max":{"version":"2.1.227","dateReleased":"2026-08-10"}},"terminal-bench-v4":{"min":{"version":"2.1.263","dateReleased":"2026-09-06"},"max":{"version":"2.1.263","dateReleased":"2026-09-06"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:c64a6ebb32778718","board":"aa_coding_agent_index","model":"gpt-6-astra","variant":"Devin Fusion CLI - GPT-6 Astra XHigh + SWE-2 Medium","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Devin Fusion CLI","reasoning_effort":"xhigh","service_mode":null,"score":58.8884,"score_low":null,"score_high":null,"mean_cost_usd_per_task":4.5365,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-astra","boardId":"aa_coding_agent_index","variantLabel":"Devin Fusion CLI - GPT-6 Astra XHigh + SWE-2 Medium","score":58.8884,"secondary":{"configurationRank":13,"agentHarness":"Devin Fusion CLI","provider":"cognition","hostModelSlug":"cognition_fusion-gpt-6-astra-xhigh-sidekick-penguin-medium","displayModel":"GPT-6 Astra XHigh + SWE-2 Medium","isUnavailable":false,"variantOf":null,"deepSWE":67.2566,"terminalBenchV4":50,"sweAtlasQnA":59.4086,"meanCostUsdPerTask":4.5365,"agentWallTimeSec":1481.896,"meanSteps":78.5512,"meanTotalTokens":6074134.495,"cacheHitRate":0.9201,"hostConfigId":"fa2595093fa640e73d072e7d1aa7df82","versions":{"deep-swe-v1.1":{"min":{"version":"3000.10.7","dateReleased":null},"max":{"version":"3000.10.7","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"3000.10.7","dateReleased":null},"max":{"version":"3000.10.7","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"3000.10.7","dateReleased":null},"max":{"version":"3000.10.7","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:79db3062d86f1bbd","board":"aa_coding_agent_index","model":"gpt-6.1-sol","variant":"Codex - GPT-6.1 Sol (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"low","service_mode":null,"score":57.2163,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.4988,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6.1-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6.1 Sol (low) ({'reasoning_effort': 'low'})","score":57.2163,"secondary":{"configurationRank":14,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_quasar-alpha","displayModel":"GPT-6.1 Sol (low)","isUnavailable":false,"variantOf":null,"deepSWE":67.5516,"terminalBenchV4":48.9899,"sweAtlasQnA":55.1075,"meanCostUsdPerTask":0.4988,"agentWallTimeSec":517.8139,"meanSteps":35.7404,"meanTotalTokens":1713818.9182,"cacheHitRate":0.9136,"hostConfigId":"fa589dad83ffcdd6cce72a467e552063","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:a5a1b714fe3ad343","board":"aa_coding_agent_index","model":"gpt-6-sol","variant":"Codex - GPT-6 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":56.6626,"score_low":null,"score_high":null,"mean_cost_usd_per_task":2.9897,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6 Sol (max)","score":56.6626,"secondary":{"configurationRank":15,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_gpt-6-sol","displayModel":"GPT-6 Sol (max)","isUnavailable":false,"variantOf":null,"deepSWE":69.0265,"terminalBenchV4":43.4343,"sweAtlasQnA":57.5269,"meanCostUsdPerTask":2.9897,"agentWallTimeSec":1338.0527,"meanSteps":86.5501,"meanTotalTokens":9840777.7635,"cacheHitRate":0.964,"hostConfigId":"7a0dfd2a7a9d93056f860d4675586d66","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:fa75b2e79a4d8cb4","board":"aa_coding_agent_index","model":"grok-4.7","variant":"Grok Build - Grok 4.7 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Grok Build","reasoning_effort":"xhigh","service_mode":null,"score":56.2676,"score_low":null,"score_high":null,"mean_cost_usd_per_task":8.8175,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.7","boardId":"aa_coding_agent_index","variantLabel":"Grok Build - Grok 4.7 (xhigh)","score":56.2676,"secondary":{"configurationRank":16,"agentHarness":"Grok Build","provider":"xai","hostModelSlug":"xai_grok-bop-o86-xhigh","displayModel":"Grok 4.7 (xhigh)","isUnavailable":false,"variantOf":null,"deepSWE":72.5664,"terminalBenchV4":33.3333,"sweAtlasQnA":62.9032,"meanCostUsdPerTask":8.8175,"agentWallTimeSec":2354.3677,"meanSteps":162.6458,"meanTotalTokens":14324724.2838,"cacheHitRate":0.9389,"hostConfigId":"f927d9c41b5c3021996eb14c3dd1341e","versions":{"deep-swe-v1.1":{"min":{"version":"1.0.40","dateReleased":null},"max":{"version":"1.0.40","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"1.0.40","dateReleased":null},"max":{"version":"1.0.40","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"1.0.40","dateReleased":null},"max":{"version":"1.0.40","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:197c477622a1763e","board":"aa_coding_agent_index","model":"claude-sonnet-5.5","variant":"Claude Code - Sonnet 5.5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"high","service_mode":null,"score":55.0125,"score_low":null,"score_high":null,"mean_cost_usd_per_task":1.2355,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Sonnet 5.5 (high)","score":55.0125,"secondary":{"configurationRank":17,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"private-anthropic_claude-beignet-eap-high","displayModel":"Sonnet 5.5 (high)","isUnavailable":false,"variantOf":null,"deepSWE":66.6667,"terminalBenchV4":41.9192,"sweAtlasQnA":56.4516,"meanCostUsdPerTask":1.2355,"agentWallTimeSec":735.181,"meanSteps":36.7723,"meanTotalTokens":2652380.5655,"cacheHitRate":0.9478,"hostConfigId":"33e80f383dc644527161435b076798ad","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:538ede13f4f0982c","board":"aa_coding_agent_index","model":"gpt-5.6-sol","variant":"Codex - GPT-5.6 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":54.5591,"score_low":null,"score_high":null,"mean_cost_usd_per_task":6.3533,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-sol","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-5.6 Sol (max) ({'reasoning_effort': 'max'})","score":54.5591,"secondary":{"configurationRank":18,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_gpt-5-6-sol","displayModel":"GPT-5.6 Sol (max)","isUnavailable":false,"variantOf":null,"deepSWE":72.2714,"terminalBenchV4":37.3737,"sweAtlasQnA":54.0323,"meanCostUsdPerTask":6.3533,"agentWallTimeSec":1237.6964,"meanSteps":116.3949,"meanTotalTokens":10170136.6326,"cacheHitRate":0.9458,"hostConfigId":"709f801275dedae5675b1f56e97c9df3","versions":{"deep-swe-v1.1":{"min":{"version":"0.153.4","dateReleased":"2026-09-04"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}},"swe-atlas-qna":{"min":{"version":"0.139.0","dateReleased":"2026-06-09"},"max":{"version":"0.147.0","dateReleased":"2026-08-07"}},"terminal-bench-v4":{"min":{"version":"0.153.4","dateReleased":"2026-09-04"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:36a45d40bcfc3714","board":"aa_coding_agent_index","model":"muse-spark-1.3","variant":"Muse Code - Muse Spark 1.3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Muse Code","reasoning_effort":"max","service_mode":null,"score":54.3027,"score_low":null,"score_high":null,"mean_cost_usd_per_task":3.9775,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.3","boardId":"aa_coding_agent_index","variantLabel":"Muse Code - Muse Spark 1.3 (max)","score":54.3027,"secondary":{"configurationRank":19,"agentHarness":"Muse Code","provider":"meta","hostModelSlug":"meta_aa_glacier135","displayModel":"Muse Spark 1.3 (max)","isUnavailable":false,"variantOf":null,"deepSWE":71.6814,"terminalBenchV4":31.8182,"sweAtlasQnA":59.4086,"meanCostUsdPerTask":3.9775,"agentWallTimeSec":1106.8534,"meanSteps":185.2382,"meanTotalTokens":16709444.5737,"cacheHitRate":0.9713,"hostConfigId":"f2de07373c3626f1b9269a6ecad7804c","versions":{"deep-swe-v1.1":{"min":{"version":"1.0.3-R2198.1","dateReleased":"2026-09-04"},"max":{"version":"1.0.3-R2198.1","dateReleased":"2026-09-04"}},"swe-atlas-qna":{"min":{"version":"1.0.3-R2198.1","dateReleased":"2026-09-04"},"max":{"version":"1.0.3-R2198.1","dateReleased":"2026-09-04"}},"terminal-bench-v4":{"min":{"version":"1.0.3-R2198.1","dateReleased":"2026-09-04"},"max":{"version":"1.0.3-R2198.1","dateReleased":"2026-09-04"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:9f41ae70311c2b0b","board":"aa_coding_agent_index","model":"glm-5.3","variant":"Opencode - GLM-5.3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Opencode","reasoning_effort":"max","service_mode":null,"score":53.5548,"score_low":null,"score_high":null,"mean_cost_usd_per_task":4.2414,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"glm-5.3","boardId":"aa_coding_agent_index","variantLabel":"Opencode - GLM-5.3 ({'reasoning_effort': 'max'})","score":53.5548,"secondary":{"configurationRank":20,"agentHarness":"Opencode","provider":"zai","hostModelSlug":"zai_glm-5-3","displayModel":"GLM-5.3","isUnavailable":false,"variantOf":null,"deepSWE":61.3569,"terminalBenchV4":39.899,"sweAtlasQnA":59.4086,"meanCostUsdPerTask":4.2414,"agentWallTimeSec":2887.9557,"meanSteps":107.0495,"meanTotalTokens":14276424.4268,"cacheHitRate":0.9746,"hostConfigId":"45493c2782f1bd94f6d30d4911f317a0","versions":{"deep-swe-v1.1":{"min":{"version":"1.18.29","dateReleased":"2026-09-04"},"max":{"version":"1.18.29","dateReleased":"2026-09-04"}},"swe-atlas-qna":{"min":{"version":"1.18.29","dateReleased":"2026-09-04"},"max":{"version":"1.18.29","dateReleased":"2026-09-04"}},"terminal-bench-v4":{"min":{"version":"1.18.29","dateReleased":"2026-09-04"},"max":{"version":"1.18.29","dateReleased":"2026-09-04"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:4a7256426ed97ee6","board":"aa_coding_agent_index","model":"kimi-k3","variant":"Kimi Code CLI - Kimi K3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Kimi Code CLI","reasoning_effort":null,"service_mode":null,"score":51.9259,"score_low":null,"score_high":null,"mean_cost_usd_per_task":5.0472,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"kimi-k3","boardId":"aa_coding_agent_index","variantLabel":"Kimi Code CLI - Kimi K3","score":51.9259,"secondary":{"configurationRank":21,"agentHarness":"Kimi Code CLI","provider":"moonshotai","hostModelSlug":"moonshot_kimi-k3","displayModel":"Kimi K3","isUnavailable":false,"variantOf":null,"deepSWE":68.4366,"terminalBenchV4":21.2121,"sweAtlasQnA":66.129,"meanCostUsdPerTask":5.0472,"agentWallTimeSec":3689.3442,"meanSteps":173.3603,"meanTotalTokens":8481121.5336,"cacheHitRate":0.95,"hostConfigId":"55b029d6971b4d378335bb168eb98b3b","versions":{"deep-swe-v1.1":{"min":{"version":"0.36.0","dateReleased":"2026-08-13"},"max":{"version":"0.36.1","dateReleased":"2026-08-14"}},"swe-atlas-qna":{"min":{"version":"0.26.0","dateReleased":"2026-07-16"},"max":{"version":"0.26.0","dateReleased":"2026-07-16"}},"terminal-bench-v4":{"min":{"version":"0.36.1","dateReleased":"2026-08-14"},"max":{"version":"0.36.1","dateReleased":"2026-08-14"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:2aeda79a0ee5dba5","board":"aa_coding_agent_index","model":"muse-spark-1.3","variant":"Muse Code - Muse Spark 1.3 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Muse Code 1.0.2 RC","reasoning_effort":"xhigh","service_mode":null,"score":48.2993,"score_low":null,"score_high":null,"mean_cost_usd_per_task":3.4678,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"muse-spark-1.3","boardId":"aa_coding_agent_index","variantLabel":"Muse Code - Muse Spark 1.3 (xhigh)","score":48.2993,"secondary":{"configurationRank":22,"agentHarness":"Muse Code 1.0.2 RC","provider":"meta","hostModelSlug":"meta_joyful-jello138","displayModel":"Muse Spark 1.3 (xhigh)","isUnavailable":false,"variantOf":null,"deepSWE":73.1563,"terminalBenchV4":17.1717,"sweAtlasQnA":54.5699,"meanCostUsdPerTask":3.4678,"agentWallTimeSec":1098.2029,"meanSteps":187.4895,"meanTotalTokens":16237094.352,"cacheHitRate":0.9662,"hostConfigId":"cd60470c35fd626a116af80fd5915dfa","versions":{"deep-swe-v1.1":{"min":{"version":"1.0.2-R2040.1","dateReleased":"2026-09-02"},"max":{"version":"1.0.2-R2040.1","dateReleased":"2026-09-02"}},"swe-atlas-qna":{"min":{"version":"1.0.2-R2040.1","dateReleased":"2026-09-02"},"max":{"version":"1.0.2-R2040.1","dateReleased":"2026-09-02"}},"terminal-bench-v4":{"min":{"version":"1.0.2-R2040.1","dateReleased":"2026-09-02"},"max":{"version":"1.0.2-R2040.1","dateReleased":"2026-09-02"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:1157285386872e0a","board":"aa_coding_agent_index","model":"grok-4.6","variant":"Grok Build - Grok 4.6 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Grok Build","reasoning_effort":"xhigh","service_mode":null,"score":46.969,"score_low":null,"score_high":null,"mean_cost_usd_per_task":3.5701,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"grok-4.6","boardId":"aa_coding_agent_index","variantLabel":"Grok Build - Grok 4.6 (xhigh)","score":46.969,"secondary":{"configurationRank":23,"agentHarness":"Grok Build","provider":"xai","hostModelSlug":"xai_grok-4-6-xhigh","displayModel":"Grok 4.6 (xhigh)","isUnavailable":false,"variantOf":null,"deepSWE":64.8968,"terminalBenchV4":17.6768,"sweAtlasQnA":58.3333,"meanCostUsdPerTask":3.5701,"agentWallTimeSec":1167.3387,"meanSteps":112.0616,"meanTotalTokens":5543624.5974,"cacheHitRate":0.9058,"hostConfigId":"70a5d300021b9986f77d4198272ef1a5","versions":{"deep-swe-v1.1":{"min":{"version":"1.0.13","dateReleased":"2026-08-28"},"max":{"version":"1.0.13","dateReleased":"2026-08-28"}},"swe-atlas-qna":{"min":{"version":"1.0.13","dateReleased":"2026-08-28"},"max":{"version":"1.0.13","dateReleased":"2026-08-28"}},"terminal-bench-v4":{"min":{"version":"1.0.13","dateReleased":"2026-08-28"},"max":{"version":"1.0.13","dateReleased":"2026-08-28"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:5b82e4f8ad6e49f3","board":"aa_coding_agent_index","model":"claude-sonnet-5.5","variant":"Claude Code - Sonnet 5.5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"medium","service_mode":null,"score":45.884,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.6188,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Sonnet 5.5 (medium)","score":45.884,"secondary":{"configurationRank":24,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"private-anthropic_claude-beignet-eap-medium","displayModel":"Sonnet 5.5 (medium)","isUnavailable":false,"variantOf":null,"deepSWE":65.4867,"terminalBenchV4":27.2727,"sweAtlasQnA":44.8925,"meanCostUsdPerTask":0.6188,"agentWallTimeSec":510.7722,"meanSteps":22.209,"meanTotalTokens":1253351.4928,"cacheHitRate":0.9341,"hostConfigId":"4d6aa9e9c8df931d54dba5956d94973c","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:82bc37b4bd5aa375","board":"aa_coding_agent_index","model":"qwen3.8-max","variant":"Claude Code - Qwen3.8 Max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":43.2653,"score_low":null,"score_high":null,"mean_cost_usd_per_task":3.479,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"qwen3.8-max","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Qwen3.8 Max","score":43.2653,"secondary":{"configurationRank":25,"agentHarness":"Claude Code","provider":"alibaba_cloud","hostModelSlug":"alibaba_cloud_qwen3-8-max-public","displayModel":"Qwen3.8 Max","isUnavailable":false,"variantOf":null,"deepSWE":51.0324,"terminalBenchV4":16.6667,"sweAtlasQnA":62.0968,"meanCostUsdPerTask":3.479,"agentWallTimeSec":3784.4161,"meanSteps":110.0367,"meanTotalTokens":8706893.7981,"cacheHitRate":0.9507,"hostConfigId":"cd4c865dc2f539b650d741fd2ebf7f2d","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.261","dateReleased":"2026-09-04"},"max":{"version":"2.1.261","dateReleased":"2026-09-04"}},"swe-atlas-qna":{"min":{"version":"2.1.224","dateReleased":"2026-08-07"},"max":{"version":"2.1.236","dateReleased":"2026-08-19"}},"terminal-bench-v4":{"min":{"version":"2.1.261","dateReleased":"2026-09-04"},"max":{"version":"2.1.261","dateReleased":"2026-09-04"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:cdca4b328af5e088","board":"aa_coding_agent_index","model":"gpt-5.6-luna","variant":"Codex - GPT-5.6 Luna (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":43.2247,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.438,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-5.6-luna","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-5.6 Luna (max)","score":43.2247,"secondary":{"configurationRank":26,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_gpt-5-6-luna","displayModel":"GPT-5.6 Luna (max)","isUnavailable":false,"variantOf":null,"deepSWE":66.3717,"terminalBenchV4":14.6465,"sweAtlasQnA":48.6559,"meanCostUsdPerTask":0.438,"agentWallTimeSec":1412.2609,"meanSteps":121.5138,"meanTotalTokens":14854524.5391,"cacheHitRate":0.9719,"hostConfigId":"b3c56ec69185d544992fd578edb75e8c","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:d18f5779019b6787","board":"aa_coding_agent_index","model":"deepseek-v4-pro","variant":"Codex - DeepSeek V4 Pro 0813 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":43.052,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.238,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-pro","boardId":"aa_coding_agent_index","variantLabel":"Codex - DeepSeek V4 Pro 0813 (max)","score":43.052,"secondary":{"configurationRank":27,"agentHarness":"Codex","provider":"deepseek","hostModelSlug":"deepseek_deepseek-v4-pro-0813","displayModel":"DeepSeek V4 Pro 0813 (max)","isUnavailable":false,"variantOf":null,"deepSWE":57.2271,"terminalBenchV4":10.101,"sweAtlasQnA":61.828,"meanCostUsdPerTask":0.238,"agentWallTimeSec":2417.55,"meanSteps":138.1463,"meanTotalTokens":24180108.0281,"cacheHitRate":0.9898,"hostConfigId":"542590ecc1aa12a718d158565d464daa","versions":{"deep-swe-v1.1":{"min":{"version":"0.153.2","dateReleased":"2026-09-03"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}},"swe-atlas-qna":{"min":{"version":"0.147.0","dateReleased":"2026-08-07"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}},"terminal-bench-v4":{"min":{"version":"0.153.2","dateReleased":"2026-09-03"},"max":{"version":"0.153.4","dateReleased":"2026-09-04"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:1f7a78464bab7071","board":"aa_coding_agent_index","model":"claude-sonnet-5.5","variant":"Claude Code - Sonnet 5.5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"low","service_mode":null,"score":42.0593,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.4831,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"claude-sonnet-5.5","boardId":"aa_coding_agent_index","variantLabel":"Claude Code - Sonnet 5.5 (low)","score":42.0593,"secondary":{"configurationRank":28,"agentHarness":"Claude Code","provider":"anthropic","hostModelSlug":"private-anthropic_claude-beignet-eap-low","displayModel":"Sonnet 5.5 (low)","isUnavailable":false,"variantOf":null,"deepSWE":61.9469,"terminalBenchV4":25.2525,"sweAtlasQnA":38.9785,"meanCostUsdPerTask":0.4831,"agentWallTimeSec":380.8562,"meanSteps":18.5787,"meanTotalTokens":977284.7701,"cacheHitRate":0.9308,"hostConfigId":"e9f9d3ae5b03aa060114dd4b3275d67d","versions":{"deep-swe-v1.1":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"2.1.280","dateReleased":null},"max":{"version":"2.1.280","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:68de21286728c605","board":"aa_coding_agent_index","model":"gemini-3.8-flash","variant":"Antigravity SDK - Gemini 3.8 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Antigravity SDK v0.1.12","reasoning_effort":"high","service_mode":null,"score":41.8632,"score_low":null,"score_high":null,"mean_cost_usd_per_task":2.4674,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gemini-3.8-flash","boardId":"aa_coding_agent_index","variantLabel":"Antigravity SDK - Gemini 3.8 Flash (high)","score":41.8632,"secondary":{"configurationRank":29,"agentHarness":"Antigravity SDK v0.1.12","provider":"google","hostModelSlug":"google_skimaki_ai-studio","displayModel":"Gemini 3.8 Flash (high)","isUnavailable":false,"variantOf":null,"deepSWE":65.7817,"terminalBenchV4":14.6465,"sweAtlasQnA":45.1613,"meanCostUsdPerTask":2.4674,"agentWallTimeSec":702.711,"meanSteps":188.4884,"meanTotalTokens":13511073.1551,"cacheHitRate":0.8697,"hostConfigId":"d1003684d8f29b95a3b14d2a7722221a","versions":{"deep-swe-v1.1":{"min":{"version":"0.1.16","dateReleased":"2026-09-02"},"max":{"version":"0.1.16","dateReleased":"2026-09-02"}},"swe-atlas-qna":{"min":{"version":"0.1.12","dateReleased":"2026-08-13"},"max":{"version":"0.1.12","dateReleased":"2026-08-13"}},"terminal-bench-v4":{"min":{"version":"0.1.16","dateReleased":"2026-09-02"},"max":{"version":"0.1.16","dateReleased":"2026-09-02"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:01e2ffcaa1665576","board":"aa_coding_agent_index","model":"gpt-6-luna","variant":"Codex - GPT-6 Luna (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":41.0744,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.1759,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"gpt-6-luna","boardId":"aa_coding_agent_index","variantLabel":"Codex - GPT-6 Luna (max) ({'reasoning_effort': 'max'})","score":41.0744,"secondary":{"configurationRank":30,"agentHarness":"Codex","provider":"openai","hostModelSlug":"openai_gpt-6-luna","displayModel":"GPT-6 Luna (max)","isUnavailable":false,"variantOf":null,"deepSWE":63.7168,"terminalBenchV4":15.1515,"sweAtlasQnA":44.3548,"meanCostUsdPerTask":0.1759,"agentWallTimeSec":1285.9269,"meanSteps":88.643,"meanTotalTokens":10218069.2481,"cacheHitRate":0.9606,"hostConfigId":"ac8276cd5d15ac2daacb9956ad44c132","versions":{"deep-swe-v1.1":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"swe-atlas-qna":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}},"terminal-bench-v4":{"min":{"version":"0.154.0","dateReleased":null},"max":{"version":"0.154.0","dateReleased":null}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"aa_coding_agent_index:77cf38495a1886a0","board":"aa_coding_agent_index","model":"deepseek-v4-flash","variant":"Codex - DeepSeek V4 Flash 0731 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":38.7425,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.0851,"median_cost_usd_per_task":null,"source":"https://artificialanalysis.ai/agents/coding-agents","checked_at":"2026-10-01","archive":"scores-aa-round5-2026-10-01.json","raw_record":{"model":"deepseek-v4-flash","boardId":"aa_coding_agent_index","variantLabel":"Codex - DeepSeek V4 Flash 0731 (max)","score":38.7425,"secondary":{"configurationRank":31,"agentHarness":"Codex","provider":"deepseek","hostModelSlug":"deepseek_deepseek-v4-flash-0731","displayModel":"DeepSeek V4 Flash 0731 (max)","isUnavailable":false,"variantOf":null,"deepSWE":54.2773,"terminalBenchV4":10.6061,"sweAtlasQnA":51.3441,"meanCostUsdPerTask":0.0851,"agentWallTimeSec":1097.0404,"meanSteps":121.6667,"meanTotalTokens":15350225.2992,"cacheHitRate":0.987,"hostConfigId":"28ef65135f08f035dc93a97d324fea73","versions":{"deep-swe-v1.1":{"min":{"version":"0.153.2","dateReleased":"2026-09-03"},"max":{"version":"0.153.2","dateReleased":"2026-09-03"}},"swe-atlas-qna":{"min":{"version":"0.146.0","dateReleased":"2026-07-29"},"max":{"version":"0.146.1","dateReleased":"2026-08-05"}},"terminal-bench-v4":{"min":{"version":"0.153.2","dateReleased":"2026-09-03"},"max":{"version":"0.153.2","dateReleased":"2026-09-03"}}}},"source":"https://artificialanalysis.ai/agents/coding-agents","checkedAt":"2026-10-01","note":"Score belongs to this named agent/harness × model configuration; it is not a harness-independent model measurement. indexScore stored as percent (raw fraction × 100)."}},{"configuration_id":"terminal_bench_4:e5bdec03718dfb8f","board":"terminal_bench_4","model":"gpt-6-astra","variant":"Codex - GPT-6 Astra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":58.18,"score_low":55.39,"score_high":60.97,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-6-astra","variantLabel":"Codex - GPT-6 Astra (max)","score":58.18,"secondary":{"configurationRank":1,"agentHarness":"Codex","reasoningEffort":"max","runDate":"2026-09-03","nTasks":66,"nTrials":330,"successes":192,"ciMinus":2.79,"ciPlus":2.79,"passAt2":0.6485,"passAt3":0.6788,"passAt4":0.697,"passAt5":0.7121,"totalCostUsd":3267.18,"costUsdPerTrialTask":9.9,"totalTokens":1529778322,"outputTokens":23988992,"cachedInputTokens":1443351702,"uncachedInputTokens":1505789330,"avgTrialDurationSec":2796.3,"entryId":"5c537be4-7fc3-449b-8bfc-ceb9061c2535","entryUpdatedAt":"2026-09-10T21:58:00.001022+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:136537b36ce98c25","board":"terminal_bench_4","model":"claude-fable-5.1","variant":"Claude Code - Fable 5.1 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":57.88,"score_low":54.120000000000005,"score_high":61.64,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-fable-5.1","variantLabel":"Claude Code - Fable 5.1 (max)","score":57.88,"secondary":{"configurationRank":2,"agentHarness":"Claude Code","reasoningEffort":"max","runDate":"2026-09-01","nTasks":66,"nTrials":330,"successes":191,"ciMinus":3.76,"ciPlus":3.76,"passAt2":0.7,"passAt3":0.747,"passAt4":0.7727,"passAt5":0.7879,"totalCostUsd":6243.5,"costUsdPerTrialTask":18.92,"totalTokens":2746221560,"outputTokens":63121568,"cachedInputTokens":2467056906,"uncachedInputTokens":2683099992,"avgTrialDurationSec":3893.5,"entryId":"c741608e-c94e-417d-b7a8-e67111a0c887","entryUpdatedAt":"2026-09-03T02:56:08.149542+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:d1970c4faffe81bc","board":"terminal_bench_4","model":"gpt-6-astra","variant":"Codex - GPT-6 Astra (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"xhigh","service_mode":null,"score":57.88,"score_low":55.160000000000004,"score_high":60.6,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-6-astra","variantLabel":"Codex - GPT-6 Astra (xhigh)","score":57.88,"secondary":{"configurationRank":2,"agentHarness":"Codex","reasoningEffort":"xhigh","runDate":"2026-09-03","nTasks":66,"nTrials":330,"successes":191,"ciMinus":2.72,"ciPlus":2.72,"passAt2":0.6424,"passAt3":0.6697,"passAt4":0.6848,"passAt5":0.697,"totalCostUsd":2350.51,"costUsdPerTrialTask":7.12,"totalTokens":1201907334,"outputTokens":14950287,"cachedInputTokens":1140730194,"uncachedInputTokens":1186957047,"avgTrialDurationSec":2208.5,"entryId":"16db8ad5-84aa-4588-b660-1ce68c0d45e2","entryUpdatedAt":"2026-09-10T21:57:57.437008+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:425d97cf71b364f9","board":"terminal_bench_4","model":"gpt-6-astra","variant":"Codex - GPT-6 Astra (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"high","service_mode":null,"score":57.88,"score_low":54.910000000000004,"score_high":60.85,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-6-astra","variantLabel":"Codex - GPT-6 Astra (high)","score":57.88,"secondary":{"configurationRank":2,"agentHarness":"Codex","reasoningEffort":"high","runDate":"2026-09-03","nTasks":66,"nTrials":330,"successes":191,"ciMinus":2.97,"ciPlus":2.97,"passAt2":0.6545,"passAt3":0.6864,"passAt4":0.703,"passAt5":0.7121,"totalCostUsd":2269.42,"costUsdPerTrialTask":6.88,"totalTokens":1208224034,"outputTokens":13703943,"cachedInputTokens":1151219647,"uncachedInputTokens":1194520091,"avgTrialDurationSec":2113.7,"entryId":"3475050c-bf5e-4261-a3f6-0af5350af13f","entryUpdatedAt":"2026-09-10T21:57:54.867871+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:cc67912665adccef","board":"terminal_bench_4","model":"claude-fable-5.1","variant":"Claude Code - Fable 5.1 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"xhigh","service_mode":null,"score":57.88,"score_low":54.52,"score_high":61.24,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-fable-5.1","variantLabel":"Claude Code - Fable 5.1 (xhigh)","score":57.88,"secondary":{"configurationRank":2,"agentHarness":"Claude Code","reasoningEffort":"xhigh","runDate":"2026-09-01","nTasks":66,"nTrials":330,"successes":191,"ciMinus":3.36,"ciPlus":3.36,"passAt2":0.6758,"passAt3":0.7152,"passAt4":0.7394,"passAt5":0.7576,"totalCostUsd":4872.04,"costUsdPerTrialTask":14.76,"totalTokens":2343414000,"outputTokens":50426931,"cachedInputTokens":2140665130,"uncachedInputTokens":2292987069,"avgTrialDurationSec":3202.1,"entryId":"9e1c5e11-748f-4f2e-911a-070a396d5bf8","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:c1be8fa31509c696","board":"terminal_bench_4","model":"claude-fable-5.1","variant":"Claude Code - Fable 5.1 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"high","service_mode":null,"score":54.55,"score_low":51.11,"score_high":57.989999999999995,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-fable-5.1","variantLabel":"Claude Code - Fable 5.1 (high)","score":54.55,"secondary":{"configurationRank":6,"agentHarness":"Claude Code","reasoningEffort":"high","runDate":"2026-09-01","nTasks":66,"nTrials":330,"successes":180,"ciMinus":3.44,"ciPlus":3.44,"passAt2":0.647,"passAt3":0.697,"passAt4":0.7303,"passAt5":0.7576,"totalCostUsd":3985.38,"costUsdPerTrialTask":12.08,"totalTokens":2233621484,"outputTokens":37429044,"cachedInputTokens":2064726577,"uncachedInputTokens":2196192440,"avgTrialDurationSec":3017.7,"entryId":"4643b23c-53e8-47ae-a3bd-9abc900e9588","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:1ebcc077b6f11845","board":"terminal_bench_4","model":"gpt-6-astra","variant":"Codex - GPT-6 Astra (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"medium","service_mode":null,"score":54.24,"score_low":51.58,"score_high":56.900000000000006,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-6-astra","variantLabel":"Codex - GPT-6 Astra (medium)","score":54.24,"secondary":{"configurationRank":7,"agentHarness":"Codex","reasoningEffort":"medium","runDate":"2026-09-03","nTasks":66,"nTrials":330,"successes":179,"ciMinus":2.66,"ciPlus":2.66,"passAt2":0.603,"passAt3":0.6318,"passAt4":0.6515,"passAt5":0.6667,"totalCostUsd":1914.8,"costUsdPerTrialTask":5.8,"totalTokens":1051885207,"outputTokens":10770605,"cachedInputTokens":1003874758,"uncachedInputTokens":1041114602,"avgTrialDurationSec":1885.9,"entryId":"f3c3d5a6-6424-4acb-bfcc-615c3f79f6dd","entryUpdatedAt":"2026-09-10T21:57:52.562258+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:5edd3a6853cd6339","board":"terminal_bench_4","model":"claude-fable-5.1","variant":"Claude Code - Fable 5.1 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"medium","service_mode":null,"score":53.94,"score_low":50.55,"score_high":57.33,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-fable-5.1","variantLabel":"Claude Code - Fable 5.1 (medium)","score":53.94,"secondary":{"configurationRank":8,"agentHarness":"Claude Code","reasoningEffort":"medium","runDate":"2026-09-01","nTasks":66,"nTrials":330,"successes":178,"ciMinus":3.39,"ciPlus":3.39,"passAt2":0.6379,"passAt3":0.6788,"passAt4":0.7,"passAt5":0.7121,"totalCostUsd":2832.9,"costUsdPerTrialTask":8.58,"totalTokens":1620460859,"outputTokens":25991800,"cachedInputTokens":1498585169,"uncachedInputTokens":1594469059,"avgTrialDurationSec":2528.3,"entryId":"83e6bf7e-aeb4-4116-9f30-ca8de96603fb","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:7d59ed14e6a26f29","board":"terminal_bench_4","model":"claude-opus-5","variant":"Claude Code - Opus 5 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"xhigh","service_mode":null,"score":53.94,"score_low":50.769999999999996,"score_high":57.11,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-5","variantLabel":"Claude Code - Opus 5 (xhigh)","score":53.94,"secondary":{"configurationRank":8,"agentHarness":"Claude Code","reasoningEffort":"xhigh","runDate":"2026-07-24","nTasks":66,"nTrials":330,"successes":178,"ciMinus":3.17,"ciPlus":3.17,"passAt2":0.6258,"passAt3":0.6621,"passAt4":0.6848,"passAt5":0.697,"totalCostUsd":6086.22,"costUsdPerTrialTask":18.44,"totalTokens":6897141993,"outputTokens":59162574,"cachedInputTokens":6631366566,"uncachedInputTokens":6837979419,"avgTrialDurationSec":4514.7,"entryId":"f7844abc-3109-4d5f-91ad-481df7b9fac2","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:cec3b7db483cca26","board":"terminal_bench_4","model":"claude-opus-5","variant":"Claude Code - Opus 5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":51.82,"score_low":48.43,"score_high":55.21,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-5","variantLabel":"Claude Code - Opus 5 (max)","score":51.82,"secondary":{"configurationRank":10,"agentHarness":"Claude Code","reasoningEffort":"max","runDate":"2026-07-24","nTasks":66,"nTrials":330,"successes":171,"ciMinus":3.39,"ciPlus":3.39,"passAt2":0.6167,"passAt3":0.6576,"passAt4":0.6818,"passAt5":0.697,"totalCostUsd":5969.11,"costUsdPerTrialTask":18.09,"totalTokens":6527147637,"outputTokens":66012156,"cachedInputTokens":6271909956,"uncachedInputTokens":6461135481,"avgTrialDurationSec":4792.8,"entryId":"d71ac3d0-da36-49cb-89d9-323136e77111","entryUpdatedAt":"2026-09-03T00:09:44.281605+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:6c09cda642c293be","board":"terminal_bench_4","model":"gpt-6-astra","variant":"Codex - GPT-6 Astra (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"low","service_mode":null,"score":50.61,"score_low":47.86,"score_high":53.36,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-6-astra","variantLabel":"Codex - GPT-6 Astra (low)","score":50.61,"secondary":{"configurationRank":11,"agentHarness":"Codex","reasoningEffort":"low","runDate":"2026-09-03","nTasks":66,"nTrials":330,"successes":167,"ciMinus":2.75,"ciPlus":2.75,"passAt2":0.5712,"passAt3":0.6015,"passAt4":0.6212,"passAt5":0.6364,"totalCostUsd":1557.3,"costUsdPerTrialTask":4.72,"totalTokens":889780710,"outputTokens":7988337,"cachedInputTokens":851115630,"uncachedInputTokens":881792373,"avgTrialDurationSec":1683.7,"entryId":"b3ad58f3-b311-4d4d-875b-3158cf0309d6","entryUpdatedAt":"2026-09-10T21:57:49.894535+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:82e30cded058a067","board":"terminal_bench_4","model":"claude-opus-5","variant":"Claude Code - Opus 5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"high","service_mode":null,"score":50.3,"score_low":46.57,"score_high":54.029999999999994,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-5","variantLabel":"Claude Code - Opus 5 (high)","score":50.3,"secondary":{"configurationRank":12,"agentHarness":"Claude Code","reasoningEffort":"high","runDate":"2026-07-24","nTasks":66,"nTrials":330,"successes":166,"ciMinus":3.73,"ciPlus":3.73,"passAt2":0.6227,"passAt3":0.6803,"passAt4":0.7182,"passAt5":0.7424,"totalCostUsd":4662.27,"costUsdPerTrialTask":14.13,"totalTokens":5482033334,"outputTokens":47599774,"cachedInputTokens":5303103991,"uncachedInputTokens":5434433560,"avgTrialDurationSec":3822.2,"entryId":"0b2908ee-3035-447c-9876-e68fecb3f413","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:314e4521dd78b2f2","board":"terminal_bench_4","model":"claude-opus-5","variant":"Claude Code - Opus 5 (medium)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"medium","service_mode":null,"score":44.85,"score_low":41.02,"score_high":48.68,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-5","variantLabel":"Claude Code - Opus 5 (medium)","score":44.85,"secondary":{"configurationRank":13,"agentHarness":"Claude Code","reasoningEffort":"medium","runDate":"2026-07-24","nTasks":66,"nTrials":330,"successes":148,"ciMinus":3.83,"ciPlus":3.83,"passAt2":0.5742,"passAt3":0.6379,"passAt4":0.6727,"passAt5":0.697,"totalCostUsd":3191.59,"costUsdPerTrialTask":9.67,"totalTokens":3668060107,"outputTokens":33478510,"cachedInputTokens":3541133339,"uncachedInputTokens":3634581597,"avgTrialDurationSec":3292.9,"entryId":"8c102a88-0789-4ce7-b503-bd39e84e157b","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:eb2874e55b9e0003","board":"terminal_bench_4","model":"claude-fable-5","variant":"Claude Code - Fable 5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":44.55,"score_low":40.699999999999996,"score_high":48.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-fable-5","variantLabel":"Claude Code - Fable 5 (max)","score":44.55,"secondary":{"configurationRank":14,"agentHarness":"Claude Code","reasoningEffort":"max","runDate":"2026-06-09","nTasks":66,"nTrials":330,"successes":147,"ciMinus":3.85,"ciPlus":3.85,"passAt2":0.5727,"passAt3":0.6318,"passAt4":0.6636,"passAt5":0.6818,"totalCostUsd":7265.01,"costUsdPerTrialTask":22.02,"totalTokens":3785220369,"outputTokens":58619905,"cachedInputTokens":3564065093,"uncachedInputTokens":3726600464,"avgTrialDurationSec":4202.8,"entryId":"36c077e0-4879-4444-b315-8532d66401d6","entryUpdatedAt":"2026-09-03T00:09:00.046808+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:4bcb6aba27442f07","board":"terminal_bench_4","model":"claude-fable-5.1","variant":"Claude Code - Fable 5.1 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"low","service_mode":null,"score":43.33,"score_low":39.72,"score_high":46.94,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-fable-5.1","variantLabel":"Claude Code - Fable 5.1 (low)","score":43.33,"secondary":{"configurationRank":15,"agentHarness":"Claude Code","reasoningEffort":"low","runDate":"2026-09-01","nTasks":66,"nTrials":330,"successes":143,"ciMinus":3.61,"ciPlus":3.61,"passAt2":0.5455,"passAt3":0.6,"passAt4":0.6303,"passAt5":0.6515,"totalCostUsd":2358.72,"costUsdPerTrialTask":7.15,"totalTokens":1334012516,"outputTokens":19935011,"cachedInputTokens":1225657226,"uncachedInputTokens":1314077505,"avgTrialDurationSec":2290.8,"entryId":"e18d02ab-0973-4d86-872f-a353c8a7f344","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:0bfd0655b217cb23","board":"terminal_bench_4","model":"glm-5.3","variant":"Claude Code - GLM-5.3 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":41.82,"score_low":38.59,"score_high":45.05,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"glm-5.3","variantLabel":"Claude Code - GLM-5.3 (max)","score":41.82,"secondary":{"configurationRank":16,"agentHarness":"Claude Code","reasoningEffort":"max","runDate":"2026-08-14","nTasks":66,"nTrials":330,"successes":138,"ciMinus":3.23,"ciPlus":3.23,"passAt2":0.5076,"passAt3":0.547,"passAt4":0.5667,"passAt5":0.5758,"totalCostUsd":2727.63,"costUsdPerTrialTask":8.27,"totalTokens":8677039373,"outputTokens":68660047,"cachedInputTokens":8444041600,"uncachedInputTokens":8608379326,"avgTrialDurationSec":5831,"entryId":"d72e8775-f4f1-4313-99e7-35b1cb499f24","entryUpdatedAt":"2026-09-03T00:08:03.780431+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:bb8c22faf493862e","board":"terminal_bench_4","model":"grok-4.7","variant":"Grok Build - Grok 4.7 (xhigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Grok Build","reasoning_effort":"xhigh","service_mode":null,"score":37.58,"score_low":34.04,"score_high":41.12,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"grok-4.7","variantLabel":"Grok Build - Grok 4.7 (xhigh)","score":37.58,"secondary":{"configurationRank":17,"agentHarness":"Grok Build","reasoningEffort":"xhigh","runDate":"2026-09-21","nTasks":66,"nTrials":330,"successes":124,"ciMinus":3.54,"ciPlus":3.54,"passAt2":0.4833,"passAt3":0.5394,"passAt4":0.5758,"passAt5":0.6061,"totalCostUsd":3683.287146,"costUsdPerTrialTask":11.16,"totalTokens":5485862389,"outputTokens":22977067,"cachedInputTokens":5207370880,"uncachedInputTokens":255514442,"avgTrialDurationSec":5700.9,"entryId":"84b39f56-fe3c-46c3-919f-3b67b73f9b49","entryUpdatedAt":"2026-09-21T21:28:10.802412+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:52ccc5571264dbe4","board":"terminal_bench_4","model":"gpt-5.6-sol","variant":"Codex - GPT-5.6 Sol (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":37.27,"score_low":33.49,"score_high":41.050000000000004,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-5.6-sol","variantLabel":"Codex - GPT-5.6 Sol (max)","score":37.27,"secondary":{"configurationRank":18,"agentHarness":"Codex","reasoningEffort":"max","runDate":"2026-06-26","nTasks":66,"nTrials":330,"successes":123,"ciMinus":3.78,"ciPlus":3.78,"passAt2":0.4955,"passAt3":0.55,"passAt4":0.5818,"passAt5":0.6061,"totalCostUsd":2541.7,"costUsdPerTrialTask":7.7,"totalTokens":4409948969,"outputTokens":23451415,"cachedInputTokens":4316454466,"uncachedInputTokens":4386497554,"avgTrialDurationSec":2388.4,"entryId":"0e349bde-b264-494d-a853-fada9c696192","entryUpdatedAt":"2026-09-03T00:09:13.502808+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:3ef9fcfb65e0cb76","board":"terminal_bench_4","model":"claude-opus-5","variant":"Claude Code - Opus 5 (low)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"low","service_mode":null,"score":34.85,"score_low":30.91,"score_high":38.79,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-5","variantLabel":"Claude Code - Opus 5 (low)","score":34.85,"secondary":{"configurationRank":19,"agentHarness":"Claude Code","reasoningEffort":"low","runDate":"2026-07-24","nTasks":66,"nTrials":330,"successes":115,"ciMinus":3.94,"ciPlus":3.94,"passAt2":0.4818,"passAt3":0.547,"passAt4":0.5879,"passAt5":0.6212,"totalCostUsd":2393.88,"costUsdPerTrialTask":7.25,"totalTokens":2721178403,"outputTokens":23835411,"cachedInputTokens":2619174445,"uncachedInputTokens":2697342992,"avgTrialDurationSec":2947.3,"entryId":"03fe4f10-b683-427b-a820-61e38145d28b","entryUpdatedAt":"2026-09-17T23:52:40.302826+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:4b4e1e5e2f86b9f3","board":"terminal_bench_4","model":"claude-opus-4.8","variant":"Claude Code - Opus 4.8 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":23.64,"score_low":20.080000000000002,"score_high":27.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-4.8","variantLabel":"Claude Code - Opus 4.8 (max)","score":23.64,"secondary":{"configurationRank":20,"agentHarness":"Claude Code","reasoningEffort":"max","runDate":"2026-05-28","nTasks":66,"nTrials":330,"successes":78,"ciMinus":3.56,"ciPlus":3.56,"passAt2":0.3455,"passAt3":0.4045,"passAt4":0.4424,"passAt5":0.4697,"totalCostUsd":6481.26,"costUsdPerTrialTask":19.64,"totalTokens":6415823126,"outputTokens":88813309,"cachedInputTokens":6136275149,"uncachedInputTokens":6327009817,"avgTrialDurationSec":5163,"entryId":"952b4217-421f-4d14-849b-9968fcb2063b","entryUpdatedAt":"2026-09-03T00:09:38.524394+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:57a5da2cde308af0","board":"terminal_bench_4","model":"gpt-5.6-terra","variant":"Codex - GPT-5.6 Terra (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":21.52,"score_low":18.27,"score_high":24.77,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-5.6-terra","variantLabel":"Codex - GPT-5.6 Terra (max)","score":21.52,"secondary":{"configurationRank":21,"agentHarness":"Codex","reasoningEffort":"max","runDate":"2026-06-26","nTasks":66,"nTrials":330,"successes":71,"ciMinus":3.25,"ciPlus":3.25,"passAt2":0.3061,"passAt3":0.3636,"passAt4":0.4061,"passAt5":0.4394,"totalCostUsd":1733.52,"costUsdPerTrialTask":5.25,"totalTokens":5683805798,"outputTokens":33224833,"cachedInputTokens":5560907776,"uncachedInputTokens":5650580965,"avgTrialDurationSec":2514.6,"entryId":"e53da412-5e92-408c-b369-c767924c7c1c","entryUpdatedAt":"2026-09-03T00:09:20.912429+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:2c30600255377a7d","board":"terminal_bench_4","model":"grok-4.6","variant":"Grok Build - Grok 4.6 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Grok Build","reasoning_effort":"high","service_mode":null,"score":20.3,"score_low":17.21,"score_high":23.39,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"grok-4.6","variantLabel":"Grok Build - Grok 4.6 (high)","score":20.3,"secondary":{"configurationRank":22,"agentHarness":"Grok Build","reasoningEffort":"high","runDate":"2026-08-12","nTasks":66,"nTrials":330,"successes":67,"ciMinus":3.09,"ciPlus":3.09,"passAt2":0.2848,"passAt3":0.3303,"passAt4":0.3636,"passAt5":0.3939,"totalCostUsd":3591.58,"costUsdPerTrialTask":10.88,"totalTokens":4001923428,"outputTokens":32618221,"cachedInputTokens":3877638528,"uncachedInputTokens":3969305207,"avgTrialDurationSec":2351.5,"entryId":"26354542-edc0-40cd-8f8d-9fe6fbe92ac3","entryUpdatedAt":"2026-09-03T00:09:32.707079+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:206ddf92cd474c5f","board":"terminal_bench_4","model":"gemini-3.8-flash","variant":"mini-SWE-agent - Gemini 3.8 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-SWE-agent","reasoning_effort":"high","service_mode":null,"score":19.09,"score_low":15.73,"score_high":22.45,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gemini-3.8-flash","variantLabel":"mini-SWE-agent - Gemini 3.8 Flash (high)","score":19.09,"secondary":{"configurationRank":23,"agentHarness":"mini-SWE-agent","reasoningEffort":"high","runDate":"2026-09-02","nTasks":66,"nTrials":330,"successes":63,"ciMinus":3.36,"ciPlus":3.36,"passAt2":0.2879,"passAt3":0.353,"passAt4":0.4,"passAt5":0.4394,"totalCostUsd":1828.77,"costUsdPerTrialTask":5.54,"totalTokens":17189729495,"outputTokens":68056484,"cachedInputTokens":16692887408,"uncachedInputTokens":17121673011,"avgTrialDurationSec":1995.1,"entryId":"06850434-507d-4dbd-b74a-09d52519ee35","entryUpdatedAt":"2026-09-03T18:03:18.272136+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:0b68373ce672d59b","board":"terminal_bench_4","model":"gpt-5.6-luna","variant":"Codex - GPT-5.6 Luna (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"max","service_mode":null,"score":17.27,"score_low":14.42,"score_high":20.12,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gpt-5.6-luna","variantLabel":"Codex - GPT-5.6 Luna (max)","score":17.27,"secondary":{"configurationRank":24,"agentHarness":"Codex","reasoningEffort":"max","runDate":"2026-06-26","nTasks":66,"nTrials":330,"successes":57,"ciMinus":2.85,"ciPlus":2.85,"passAt2":0.2424,"passAt3":0.2848,"passAt4":0.3121,"passAt5":0.3333,"totalCostUsd":346.67,"costUsdPerTrialTask":1.05,"totalTokens":11557463640,"outputTokens":61282672,"cachedInputTokens":11305212851,"uncachedInputTokens":11496180968,"avgTrialDurationSec":4088.3,"entryId":"51c6d76e-5baa-48c1-97b3-88656c620eef","entryUpdatedAt":"2026-09-03T00:09:06.717559+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:98d088ff379b956d","board":"terminal_bench_4","model":"grok-4.5","variant":"Grok Build - Grok 4.5 (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Grok Build","reasoning_effort":"high","service_mode":null,"score":12.42,"score_low":9.8,"score_high":15.04,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"grok-4.5","variantLabel":"Grok Build - Grok 4.5 (high)","score":12.42,"secondary":{"configurationRank":25,"agentHarness":"Grok Build","reasoningEffort":"high","runDate":"2026-07-16","nTasks":66,"nTrials":330,"successes":41,"ciMinus":2.62,"ciPlus":2.62,"passAt2":0.1833,"passAt3":0.2227,"passAt4":0.2515,"passAt5":0.2727,"totalCostUsd":2094.11,"costUsdPerTrialTask":6.35,"totalTokens":3402536479,"outputTokens":32389328,"cachedInputTokens":3248613376,"uncachedInputTokens":3370147151,"avgTrialDurationSec":3162.1,"entryId":"0eed5a0d-96b6-491b-aeb7-cd5e4700c2d7","entryUpdatedAt":"2026-09-03T00:09:26.760784+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:5718a86e9278cdf2","board":"terminal_bench_4","model":"claude-sonnet-5","variant":"Claude Code - Sonnet 5 (max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Claude Code","reasoning_effort":"max","service_mode":null,"score":12.42,"score_low":9.36,"score_high":15.48,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-sonnet-5","variantLabel":"Claude Code - Sonnet 5 (max)","score":12.42,"secondary":{"configurationRank":25,"agentHarness":"Claude Code","reasoningEffort":"max","runDate":"2026-06-30","nTasks":66,"nTrials":330,"successes":41,"ciMinus":3.06,"ciPlus":3.06,"passAt2":0.2045,"passAt3":0.2621,"passAt4":0.3091,"passAt5":0.3485,"totalCostUsd":9603.86,"costUsdPerTrialTask":29.1,"totalTokens":21562180742,"outputTokens":116598866,"cachedInputTokens":21027162464,"uncachedInputTokens":21445581876,"avgTrialDurationSec":6510.3,"entryId":"8180b9e4-4990-43d0-b906-cd8b43adaaa2","entryUpdatedAt":"2026-09-03T00:09:50.318045+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:2106790cd4627629","board":"terminal_bench_4","model":"gemini-3.7-flash","variant":"mini-SWE-agent - Gemini 3.7 Flash (high)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"mini-SWE-agent","reasoning_effort":"high","service_mode":null,"score":11.21,"score_low":8.760000000000002,"score_high":13.66,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.tbench.ai/leaderboard","checked_at":"2026-10-01","archive":"scores-terminal-bench4-round3-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"gemini-3.7-flash","variantLabel":"mini-SWE-agent - Gemini 3.7 Flash (high)","score":11.21,"secondary":{"configurationRank":27,"agentHarness":"mini-SWE-agent","reasoningEffort":"high","runDate":"2026-08-13","nTasks":66,"nTrials":330,"successes":37,"ciMinus":2.45,"ciPlus":2.45,"passAt2":0.1636,"passAt3":0.2,"passAt4":0.2303,"passAt5":0.2576,"totalCostUsd":1261.87,"costUsdPerTrialTask":3.82,"totalTokens":11137189802,"outputTokens":52134001,"cachedInputTokens":10736925351,"uncachedInputTokens":11085055801,"avgTrialDurationSec":1688.9,"entryId":"14f4da14-b86a-4dca-92d7-11178d4fe064","entryUpdatedAt":"2026-09-03T02:56:10.037161+00:00"},"source":"https://www.tbench.ai/leaderboard","checkedAt":"2026-10-01","note":"Official Terminal-Bench 4.0 leaderboard row; score belongs to this harness × model × effort configuration."}},{"configuration_id":"terminal_bench_4:b14874c72f248246","board":"terminal_bench_4","model":"deepseek-v4.1-flash","variant":"DeepSeek Harness Minimal - DeepSeek V4.1 Flash (max) [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"DeepSeek Harness Minimal","reasoning_effort":"max","service_mode":null,"score":31.2,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","checked_at":"2026-09-12","archive":"scores-terminal-bench4-selfreport-carry-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"deepseek-v4.1-flash","variantLabel":"DeepSeek Harness Minimal - DeepSeek V4.1 Flash (max)","score":31.2,"secondary":{"selfReported":true,"reportedBy":"DeepSeek","agentHarness":"DeepSeek Harness Minimal","reasoningEffort":"max","contextWindowTokens":1000000,"temperature":1,"topP":0.95},"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","checkedAt":"2026-09-12","carriedFrom":"scores-terminal-bench4-round2-selfreport-2026-09-12.json"}},{"configuration_id":"terminal_bench_4:1852bc281e1867f4","board":"terminal_bench_4","model":"swe-2","variant":"Devin CLI - SWE-2 (best effort) [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"Devin CLI","reasoning_effort":null,"service_mode":null,"score":27.3,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://cognition.com/blog/swe-2","checked_at":"2026-09-12","archive":"scores-terminal-bench4-selfreport-carry-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"swe-2","variantLabel":"Devin CLI - SWE-2 (best effort)","score":27.3,"secondary":{"selfReported":true,"reportedBy":"Cognition","agentHarness":"Devin CLI","reasoningEffort":null,"effortNote":"博客附录 A：报各 reasoning-effort 中最高分，未指明是 medium/high/max 中哪一档"},"source":"https://cognition.com/blog/swe-2","checkedAt":"2026-09-12","note":"厂商自报，非 tbench.ai 官方榜运行。Cognition 博客 Coding benchmark results 表 Terminal-Bench 4 列 27.3%。同表 Astra/Sol/Grok 4.6 与官方榜一致，Fable 5.1 55.8 与官方榜 57.88 不一致。无 CI、无成本/token 数据。见 devin-swe2-round1-2026-09-12.json。","carriedFrom":"scores-terminal-bench4-round2-selfreport-2026-09-12.json"}},{"configuration_id":"terminal_bench_4:7939bf883e242fa9","board":"terminal_bench_4","model":"claude-opus-5.5","variant":"Claude Code - Claude Opus 5.5 (default) - self-reported [vendor self-report]","score_is_estimated":null,"score_is_self_reported":true,"agent_harness":"Claude Code","reasoning_effort":null,"service_mode":null,"score":66.4,"score_low":null,"score_high":null,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://www.anthropic.com/claude-opus-5-5","checked_at":"2026-09-23","archive":"scores-terminal-bench4-selfreport-carry-2026-10-01.json","raw_record":{"boardId":"terminal_bench_4","model":"claude-opus-5.5","variantLabel":"Claude Code - Claude Opus 5.5 (default) - self-reported","score":66.4,"secondary":{"selfReported":true,"reportedBy":"Anthropic","agentHarness":"Claude Code","reasoningEffort":null,"effortNote":"发布页评测表标 at default effort，未指明是 low/medium/high/xhigh 中哪一档"},"source":"https://www.anthropic.com/claude-opus-5-5","checkedAt":"2026-09-23","note":"厂商自报，非 tbench.ai 官方榜运行。Anthropic 2026-09-22 Opus 5.5 发布页评测表 Terminal-Bench 4 行 66.4%（default effort）；同表 FrontierCode 54.4%、CursorBench 52.5%、GDPval-AA 1846 非本仓库榜单仅留证据；AA/arena/OpenDesign 暂无第三方快照分不补造。见 claude-opus55-round1-2026-09-23.json。","carriedFrom":"scores-opus55-selfreport-2026-09-23.json"}},{"configuration_id":"arena_code:1eb4c01b2d86e5d0","board":"arena_code","model":"claude-opus-5.5","variant":"claude-opus-5.5-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1818,"score_low":1801.5,"score_high":1834.5,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-opus-5.5","variantLabel":"claude-opus-5.5-max","score":1818,"secondary":{"rank":1,"rankSpread":[1,1],"ciPlus":16.5,"ciMinus":16.5,"ratingRaw":1817.8444420683181,"votes":1976,"modelKey":"claude-opus-5.5-max-vertex-webdev","modelOrganization":"Anthropic","pricePerM":"$4 / $20","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:2fa5b191581aaf87","board":"arena_code","model":"gpt-6-astra","variant":"gpt-6-astra-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1789,"score_low":1778.6,"score_high":1799.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-6-astra","variantLabel":"gpt-6-astra-max","score":1789,"secondary":{"rank":2,"rankSpread":[2,2],"ciPlus":10.4,"ciMinus":10.4,"ratingRaw":1789.1055739079766,"votes":5918,"modelKey":"gpt-6-astra-max-code-codex-harness","modelOrganization":"OpenAI","pricePerM":"$10 / $50","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:8cc100910da385f3","board":"arena_code","model":"gpt-6.1-sol","variant":"gpt-6.1-sol-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1759,"score_low":1739.9,"score_high":1778.1,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-6.1-sol","variantLabel":"gpt-6.1-sol-max","score":1759,"secondary":{"rank":3,"rankSpread":[3,4],"ciPlus":19.1,"ciMinus":19.1,"ratingRaw":1758.724697016325,"votes":1264,"modelKey":"gpt-6.1-sol-max-code-codex-harness","modelOrganization":"OpenAI","pricePerM":"$2 / $10","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:8e70472a84bf71d1","board":"arena_code","model":"claude-fable-5.1","variant":"claude-fable-5.1-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1751,"score_low":1741.3,"score_high":1760.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-fable-5.1","variantLabel":"claude-fable-5.1-max","score":1751,"secondary":{"rank":4,"rankSpread":[3,4],"ciPlus":9.7,"ciMinus":9.7,"ratingRaw":1750.6666954444622,"votes":6137,"modelKey":"claude-fable-5.1-max-webdev","modelOrganization":"Anthropic","pricePerM":"$10 / $50","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:c72d3ba206f54fd5","board":"arena_code","model":"claude-sonnet-5.5","variant":"claude-sonnet-5.5-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1709,"score_low":1693.4,"score_high":1724.6,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-sonnet-5.5","variantLabel":"claude-sonnet-5.5-high","score":1709,"secondary":{"rank":5,"rankSpread":[5,7],"ciPlus":15.6,"ciMinus":15.6,"ratingRaw":1708.6633845070737,"votes":1810,"modelKey":"claude-sonnet-5-5-high","modelOrganization":"Anthropic","pricePerM":"$2 / $10","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:3aa3e36faa602970","board":"arena_code","model":"claude-opus-5","variant":"claude-opus-5-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1694,"score_low":1687.5,"score_high":1700.5,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-opus-5","variantLabel":"claude-opus-5-max","score":1694,"secondary":{"rank":6,"rankSpread":[5,8],"ciPlus":6.5,"ciMinus":6.5,"ratingRaw":1693.678694932812,"votes":16747,"modelKey":"claude-opus-5-max-webdev","modelOrganization":"Anthropic","pricePerM":"$5 / $25","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:99658ae7db6b17a4","board":"arena_code","model":"gpt-6-sol","variant":"gpt-6-sol-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1689,"score_low":1677.2,"score_high":1700.8,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-6-sol","variantLabel":"gpt-6-sol-max","score":1689,"secondary":{"rank":7,"rankSpread":[5,10],"ciPlus":11.8,"ciMinus":11.8,"ratingRaw":1689.331621450443,"votes":3197,"modelKey":"gpt-6-sol-max-code-codex-harness","modelOrganization":"OpenAI","pricePerM":"$2 / $10","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:ab7cf4e12f67d532","board":"arena_code","model":"qwen3.8-max","variant":"qwen3.8-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1671,"score_low":1659,"score_high":1683,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"qwen3.8-max","variantLabel":"qwen3.8-max","score":1671,"secondary":{"rank":9,"rankSpread":[7,13],"ciPlus":12,"ciMinus":12,"ratingRaw":1671.2286513300737,"votes":3453,"modelKey":"kinsley-5jmg","modelOrganization":"Alibaba","pricePerM":"$1.69 / $5.07","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:8f7b0b66ed443b08","board":"arena_code","model":"qwen3.8-max-0902","variant":"qwen3.8-max-0902","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1670,"score_low":1662,"score_high":1678,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"qwen3.8-max-0902","variantLabel":"qwen3.8-max-0902","score":1670,"secondary":{"rank":10,"rankSpread":[7,13],"ciPlus":8,"ciMinus":8,"ratingRaw":1669.7153651706149,"votes":8912,"modelKey":"kiana-8mgg","modelOrganization":"Alibaba","pricePerM":"$2 / $6","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:6b8714542169429d","board":"arena_code","model":"claude-opus-5","variant":"claude-opus-5-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1660,"score_low":1654.3,"score_high":1665.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-opus-5","variantLabel":"claude-opus-5-high","score":1660,"secondary":{"rank":11,"rankSpread":[8,13],"ciPlus":5.7,"ciMinus":5.7,"ratingRaw":1659.9669958874742,"votes":21088,"modelKey":"claude-opus-5-high-webdev","modelOrganization":"Anthropic","pricePerM":"$5 / $25","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:ac2bf8a87d9de607","board":"arena_code","model":"kimi-k3","variant":"kimi-k3-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1658,"score_low":1651.6,"score_high":1664.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"kimi-k3","variantLabel":"kimi-k3-max","score":1658,"secondary":{"rank":12,"rankSpread":[9,13],"ciPlus":6.4,"ciMinus":6.4,"ratingRaw":1657.8428028964095,"votes":16246,"modelKey":"kivine-wxzc","modelOrganization":"Moonshot","pricePerM":"$3 / $15","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:d79805eb17b0ddce","board":"arena_code","model":"muse-spark-1.3","variant":"muse-spark-1.3-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1655,"score_low":1646.3,"score_high":1663.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"muse-spark-1.3","variantLabel":"muse-spark-1.3-max","score":1655,"secondary":{"rank":13,"rankSpread":[9,14],"ciPlus":8.7,"ciMinus":8.7,"ratingRaw":1655.4226502573888,"votes":7004,"modelKey":"muse-spark-1-3-max-bhma-webdev","modelOrganization":"Meta","pricePerM":"$1.25 / $4.25","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:168968cc57cebb7d","board":"arena_code","model":"grok-4.7","variant":"grok-4.7-xhigh","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":1636,"score_low":1624,"score_high":1648,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"grok-4.7","variantLabel":"grok-4.7-xhigh","score":1636,"secondary":{"rank":15,"rankSpread":[13,23],"ciPlus":12,"ciMinus":12,"ratingRaw":1636.2737652493324,"votes":3062,"modelKey":"grok-4.7-xhigh-public","modelOrganization":"SpaceXAI","pricePerM":"$2 / $6","contextLength":500000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:a9cb90573e973a11","board":"arena_code","model":"hy4-preview","variant":"hy4-preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1633,"score_low":1623.6,"score_high":1642.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"hy4-preview","variantLabel":"hy4-preview","score":1633,"secondary":{"rank":16,"rankSpread":[14,23],"ciPlus":9.4,"ciMinus":9.4,"ratingRaw":1633.1704260699548,"votes":5088,"modelKey":"phoenix-qcs2-webdev","modelOrganization":"Tencent","pricePerM":"$0.834 / $2.501","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:72ebbd507122c58e","board":"arena_code","model":"claude-fable-5","variant":"claude-fable-5-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1626,"score_low":1619.6,"score_high":1632.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-fable-5","variantLabel":"claude-fable-5-high","score":1626,"secondary":{"rank":17,"rankSpread":[14,24],"ciPlus":6.4,"ciMinus":6.4,"ratingRaw":1625.6133348674134,"votes":13952,"modelKey":"claude-fable-5-webdev","modelOrganization":"Anthropic","pricePerM":"$10 / $50","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:795cc8453d3a1a24","board":"arena_code","model":"muse-spark-1.3","variant":"muse-spark-1.3 (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":1623,"score_low":1614.2,"score_high":1631.8,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"muse-spark-1.3","variantLabel":"muse-spark-1.3 (xHigh)","score":1623,"secondary":{"rank":18,"rankSpread":[14,24],"ciPlus":8.8,"ciMinus":8.8,"ratingRaw":1623.3061076847457,"votes":6078,"modelKey":"muse-spark-1.3-xhigh-webdev","modelOrganization":"Meta","pricePerM":"$1.25 / $4.25","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:2d279d541bee2ba7","board":"arena_code","model":"glm-5.3","variant":"glm-5.3-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1622,"score_low":1613.8,"score_high":1630.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"glm-5.3","variantLabel":"glm-5.3-max","score":1622,"secondary":{"rank":19,"rankSpread":[14,24],"ciPlus":8.2,"ciMinus":8.2,"ratingRaw":1622.0411323854364,"votes":7457,"modelKey":"glm-5.3","modelOrganization":"Z.ai","pricePerM":"$1.4 / $4.4","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:dec9c8da0773c02d","board":"arena_code","model":"grok-4.6","variant":"grok-4.6-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1620,"score_low":1612.1,"score_high":1627.9,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"grok-4.6","variantLabel":"grok-4.6-high","score":1620,"secondary":{"rank":20,"rankSpread":[15,24],"ciPlus":7.9,"ciMinus":7.9,"ratingRaw":1620.0702861826376,"votes":8120,"modelKey":"grok-4.6-high-public-webdev","modelOrganization":"SpaceXAI","pricePerM":"$2 / $6","contextLength":500000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:4e5d5fbc5aef2e52","board":"arena_code","model":"deepseek-v4.1-flash","variant":"deepseek-v4.1-flash-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1620,"score_low":1609.6,"score_high":1630.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"deepseek-v4.1-flash","variantLabel":"deepseek-v4.1-flash-max","score":1620,"secondary":{"rank":21,"rankSpread":[14,25],"ciPlus":10.4,"ciMinus":10.4,"ratingRaw":1620.0078178383633,"votes":3958,"modelKey":"deepseek-v4.1-flash-max-20260910-webdev","modelOrganization":"DeepSeek","pricePerM":"$0.3 / $1.2","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:7e9fde485c6ddc62","board":"arena_code","model":"gpt-5.6-sol","variant":"gpt-5.6-sol-xhigh (codex-harness)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"xhigh","service_mode":null,"score":1619,"score_low":1613,"score_high":1625,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-5.6-sol","variantLabel":"gpt-5.6-sol-xhigh (codex-harness)","score":1619,"secondary":{"rank":22,"rankSpread":[15,24],"ciPlus":6,"ciMinus":6,"ratingRaw":1619.2947911367423,"votes":16911,"modelKey":"gpt-5.6-sol-xhigh-code-codex-harness-webdev","modelOrganization":"OpenAI","pricePerM":"$4 / $20","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:8f9ded4c04cd93a8","board":"arena_code","model":"mimo-v2.6-pro","variant":"mimo-v2.6-pro","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1619,"score_low":1606.8,"score_high":1631.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"mimo-v2.6-pro","variantLabel":"mimo-v2.6-pro","score":1619,"secondary":{"rank":23,"rankSpread":[14,25],"ciPlus":12.2,"ciMinus":12.2,"ratingRaw":1618.5497526758245,"votes":2736,"modelKey":"mimo-v2-6-pro-mw0j-webdev","modelOrganization":"Xiaomi","pricePerM":"$0.435 / $0.87","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:379b30f1d5942d86","board":"arena_code","model":"glm-5.3-flash","variant":"glm-5.3-flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1615,"score_low":1607.3,"score_high":1622.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"glm-5.3-flash","variantLabel":"glm-5.3-flash","score":1615,"secondary":{"rank":24,"rankSpread":[17,25],"ciPlus":7.7,"ciMinus":7.7,"ratingRaw":1614.7991616587792,"votes":9648,"modelKey":"glm-5.3-flash-webdev","modelOrganization":"Z.ai","pricePerM":"$0.06 / $0.2","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:828ee7654a1873aa","board":"arena_code","model":"glm-5.2","variant":"glm-5.2-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1605,"score_low":1598.6,"score_high":1611.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"glm-5.2","variantLabel":"glm-5.2-max","score":1605,"secondary":{"rank":25,"rankSpread":[22,26],"ciPlus":6.4,"ciMinus":6.4,"ratingRaw":1604.6080847709095,"votes":14543,"modelKey":"glm-5.2-code","modelOrganization":"Z.ai","pricePerM":"$1.4 / $4.4","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:778f5c97eb827e03","board":"arena_code","model":"gemini-3.7-flash","variant":"gemini-3.7-flash-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1592,"score_low":1584.6,"score_high":1599.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gemini-3.7-flash","variantLabel":"gemini-3.7-flash-high","score":1592,"secondary":{"rank":26,"rankSpread":[25,31],"ciPlus":7.4,"ciMinus":7.4,"ratingRaw":1592.2272063095438,"votes":9386,"modelKey":"andwise-evfd","modelOrganization":"Google","pricePerM":"$0.75 / $3.75","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:b8dc0ef18e30c1dc","board":"arena_code","model":"qwen3.8-27b","variant":"qwen3.8-27b","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1590,"score_low":1583,"score_high":1597,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"qwen3.8-27b","variantLabel":"qwen3.8-27b","score":1590,"secondary":{"rank":27,"rankSpread":[26,31],"ciPlus":7,"ciMinus":7,"ratingRaw":1590.4479629777084,"votes":12030,"modelKey":"qwen3.8-27b-code-webdev","modelOrganization":"Alibaba","pricePerM":"$0.4 / $3","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:cee78179b654f666","board":"arena_code","model":"gemini-3.8-flash","variant":"gemini-3.8-flash-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1583,"score_low":1575.1,"score_high":1590.9,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gemini-3.8-flash","variantLabel":"gemini-3.8-flash-high","score":1583,"secondary":{"rank":28,"rankSpread":[26,32],"ciPlus":7.9,"ciMinus":7.9,"ratingRaw":1582.6568991052472,"votes":8357,"modelKey":"skimaki-kwc6","modelOrganization":"Google","pricePerM":"$0.75 / $3.75","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:986ec1bc7da59bd9","board":"arena_code","model":"deepseek-v4-pro","variant":"deepseek-v4-pro-high-20260813","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1582,"score_low":1572.7,"score_high":1591.3,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"deepseek-v4-pro","variantLabel":"deepseek-v4-pro-high-20260813","score":1582,"secondary":{"rank":29,"rankSpread":[26,32],"ciPlus":9.3,"ciMinus":9.3,"ratingRaw":1582.3034429841264,"votes":4884,"modelKey":"deepseek-v4-pro-high-20260813-webdev","modelOrganization":"DeepSeek","pricePerM":"$1.32 / $3.96","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:27376aa104f48f0e","board":"arena_code","model":"gpt-6-luna","variant":"gpt-6-luna-max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1582,"score_low":1573,"score_high":1591,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-6-luna","variantLabel":"gpt-6-luna-max","score":1582,"secondary":{"rank":30,"rankSpread":[26,32],"ciPlus":9,"ciMinus":9,"ratingRaw":1581.9456769777462,"votes":5577,"modelKey":"gpt-6-luna-max-code-codex-harness","modelOrganization":"OpenAI","pricePerM":"$0.1 / $0.5","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:672fd54eff364145","board":"arena_code","model":"deepseek-v4-flash","variant":"deepseek-v4-flash-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1581,"score_low":1571.8,"score_high":1590.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"deepseek-v4-flash","variantLabel":"deepseek-v4-flash-high","score":1581,"secondary":{"rank":31,"rankSpread":[26,32],"ciPlus":9.2,"ciMinus":9.2,"ratingRaw":1580.9961984983802,"votes":5040,"modelKey":"deepseek-v4-flash-high-20260730-webdev","modelOrganization":"DeepSeek","pricePerM":"$0.3 / $1.2","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:5ebff85aa1280e1f","board":"arena_code","model":"step-5-preview","variant":"step-5-preview-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1570,"score_low":1556.8,"score_high":1583.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"step-5-preview","variantLabel":"step-5-preview-high","score":1570,"secondary":{"rank":32,"rankSpread":[28,36],"ciPlus":13.2,"ciMinus":13.2,"ratingRaw":1569.6312750677412,"votes":2326,"modelKey":"step-5-preview-webdev","modelOrganization":"StepFun","pricePerM":null,"contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:fae8bb30a0d7476b","board":"arena_code","model":"claude-opus-4.8","variant":"claude-opus-4-8-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1556,"score_low":1550.3,"score_high":1561.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-opus-4.8","variantLabel":"claude-opus-4-8-high","score":1556,"secondary":{"rank":35,"rankSpread":[32,37],"ciPlus":5.7,"ciMinus":5.7,"ratingRaw":1556.0210478544263,"votes":18685,"modelKey":"claude-opus-4-8-thinking-webdev","modelOrganization":"Anthropic","pricePerM":"$5 / $25","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:dc0921a30ecddd2c","board":"arena_code","model":"grok-4.5","variant":"grok-4.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1552,"score_low":1545.4,"score_high":1558.6,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"grok-4.5","variantLabel":"grok-4.5","score":1552,"secondary":{"rank":36,"rankSpread":[32,40],"ciPlus":6.6,"ciMinus":6.6,"ratingRaw":1551.9557323096842,"votes":11397,"modelKey":"grok-4.5-webdev","modelOrganization":"SpaceXAI","pricePerM":"$2 / $6","contextLength":500000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:c91ce134bd5ec2e3","board":"arena_code","model":"claude-sonnet-5","variant":"claude-sonnet-5-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1540,"score_low":1533.8,"score_high":1546.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-sonnet-5","variantLabel":"claude-sonnet-5-high","score":1540,"secondary":{"rank":39,"rankSpread":[36,43],"ciPlus":6.2,"ciMinus":6.2,"ratingRaw":1539.5183408122966,"votes":13915,"modelKey":"claude-sonnet-5-webdev","modelOrganization":"Anthropic","pricePerM":"$2 / $10","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:4b88e3c88227052e","board":"arena_code","model":"gemini-3.6-flash","variant":"gemini-3.6-flash-high","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1536,"score_low":1528.5,"score_high":1543.5,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gemini-3.6-flash","variantLabel":"gemini-3.6-flash-high","score":1536,"secondary":{"rank":41,"rankSpread":[37,43],"ciPlus":7.5,"ciMinus":7.5,"ratingRaw":1536.3195906406213,"votes":8826,"modelKey":"gemini-3.6-flash","modelOrganization":"Google","pricePerM":"$0.75 / $3.75","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:00f9bf3b98edb73a","board":"arena_code","model":"claude-opus-4.8","variant":"claude-opus-4-8","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1534,"score_low":1528.3,"score_high":1539.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"claude-opus-4.8","variantLabel":"claude-opus-4-8","score":1534,"secondary":{"rank":42,"rankSpread":[38,43],"ciPlus":5.7,"ciMinus":5.7,"ratingRaw":1534.1129659508788,"votes":17646,"modelKey":"claude-opus-4-8-webdev","modelOrganization":"Anthropic","pricePerM":"$5 / $25","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:11c91ee348aca987","board":"arena_code","model":"muse-spark-1.2","variant":"muse-spark-1.2 (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":1532,"score_low":1518.4,"score_high":1545.6,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"muse-spark-1.2","variantLabel":"muse-spark-1.2 (xHigh)","score":1532,"secondary":{"rank":43,"rankSpread":[36,48],"ciPlus":13.6,"ciMinus":13.6,"ratingRaw":1531.803564232102,"votes":2183,"modelKey":"muse-spark-1.2-xhigh-webdev","modelOrganization":"Meta","pricePerM":"$1.25 / $4.25","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:c67b556fb4592ef2","board":"arena_code","model":"gpt-5.6-terra","variant":"gpt-5.6-terra-xhigh (codex-harness)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"xhigh","service_mode":null,"score":1519,"score_low":1512.4,"score_high":1525.6,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-5.6-terra","variantLabel":"gpt-5.6-terra-xhigh (codex-harness)","score":1519,"secondary":{"rank":45,"rankSpread":[43,52],"ciPlus":6.6,"ciMinus":6.6,"ratingRaw":1519.4802448593791,"votes":12298,"modelKey":"gpt-5.6-terra-xhigh-code-codex-harness-webdev","modelOrganization":"OpenAI","pricePerM":"$2 / $12","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:fbd75732eebb198a","board":"arena_code","model":"gpt-5.6-luna","variant":"gpt-5.6-luna-xhigh (codex-harness)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"xhigh","service_mode":null,"score":1518,"score_low":1511.6,"score_high":1524.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-5.6-luna","variantLabel":"gpt-5.6-luna-xhigh (codex-harness)","score":1518,"secondary":{"rank":46,"rankSpread":[43,52],"ciPlus":6.4,"ciMinus":6.4,"ratingRaw":1517.8943144462482,"votes":12545,"modelKey":"gpt-5.6-luna-xhigh-code-codex-harness-webdev","modelOrganization":"OpenAI","pricePerM":"$0.2 / $1.2","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:ab16d803c029884c","board":"arena_code","model":"qwen3.7-max","variant":"qwen3.7-max-20260517","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1515,"score_low":1507.6,"score_high":1522.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"qwen3.7-max","variantLabel":"qwen3.7-max-20260517","score":1515,"secondary":{"rank":48,"rankSpread":[43,52],"ciPlus":7.4,"ciMinus":7.4,"ratingRaw":1514.8868189395325,"votes":8917,"modelKey":"korin-6hkw","modelOrganization":"Alibaba","pricePerM":"$1.475 / $4.425","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:8bba805c5e778260","board":"arena_code","model":"gpt-5.5","variant":"gpt-5.5-xhigh (codex-harness)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"xhigh","service_mode":null,"score":1512,"score_low":1506.1,"score_high":1517.9,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-5.5","variantLabel":"gpt-5.5-xhigh (codex-harness)","score":1512,"secondary":{"rank":49,"rankSpread":[44,53],"ciPlus":5.9,"ciMinus":5.9,"ratingRaw":1512.1472445099448,"votes":15857,"modelKey":"gpt-5.5-xhigh-webdev","modelOrganization":"OpenAI","pricePerM":"$5 / $30","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:b806900ce18f190d","board":"arena_code","model":"kimi-k2.6","variant":"kimi-k2.6","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1509,"score_low":1501.8,"score_high":1516.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"kimi-k2.6","variantLabel":"kimi-k2.6","score":1509,"secondary":{"rank":50,"rankSpread":[45,54],"ciPlus":7.2,"ciMinus":7.2,"ratingRaw":1508.7417130562972,"votes":10202,"modelKey":"kimi-k2.6-code","modelOrganization":"Moonshot","pricePerM":"$0.95 / $4","contextLength":262144},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:19874ad7a301d76d","board":"arena_code","model":"glm-5.1","variant":"glm-5.1","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1508,"score_low":1501.2,"score_high":1514.8,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"glm-5.1","variantLabel":"glm-5.1","score":1508,"secondary":{"rank":51,"rankSpread":[45,53],"ciPlus":6.8,"ciMinus":6.8,"ratingRaw":1508.458962430821,"votes":10799,"modelKey":"glm-5.1-code","modelOrganization":"Z.ai","pricePerM":"$1.4 / $4.4","contextLength":202752},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:c23c1b85207e0226","board":"arena_code","model":"hy3","variant":"hy3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1508,"score_low":1499.3,"score_high":1516.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"hy3","variantLabel":"hy3","score":1508,"secondary":{"rank":52,"rankSpread":[44,54],"ciPlus":8.7,"ciMinus":8.7,"ratingRaw":1508.107946282852,"votes":5713,"modelKey":"hy3-tencent-cloud-code","modelOrganization":"Tencent","pricePerM":"$0.132 / $0.528","contextLength":262144},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:debea18029954afc","board":"arena_code","model":"gpt-5.5","variant":"gpt-5.5-high (codex-harness)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":"high","service_mode":null,"score":1487,"score_low":1481.1,"score_high":1492.9,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-5.5","variantLabel":"gpt-5.5-high (codex-harness)","score":1487,"secondary":{"rank":56,"rankSpread":[53,61],"ciPlus":5.9,"ciMinus":5.9,"ratingRaw":1486.9649099385529,"votes":18157,"modelKey":"gpt-5.5-high-webdev","modelOrganization":"OpenAI","pricePerM":"$5 / $30","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:f206c3043bc8027f","board":"arena_code","model":"minimax-m3","variant":"minimax-m3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1482,"score_low":1476.3,"score_high":1487.7,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"minimax-m3","variantLabel":"minimax-m3","score":1482,"secondary":{"rank":57,"rankSpread":[54,62],"ciPlus":5.7,"ciMinus":5.7,"ratingRaw":1482.3817058766267,"votes":17427,"modelKey":"minimax-m3","modelOrganization":"MiniMax","pricePerM":"$0.3 / $1.2","contextLength":524288},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:efbac4e58b06eb10","board":"arena_code","model":"qwen3.6-max-preview","variant":"qwen3.6-max-preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1482,"score_low":1469.5,"score_high":1494.5,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"qwen3.6-max-preview","variantLabel":"qwen3.6-max-preview","score":1482,"secondary":{"rank":58,"rankSpread":[53,64],"ciPlus":12.5,"ciMinus":12.5,"ratingRaw":1481.814310693423,"votes":2708,"modelKey":"qwen3.6-max-preview-code","modelOrganization":"Alibaba","pricePerM":"$1.027 / $6.162","contextLength":262144},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:9e023543b5154b09","board":"arena_code","model":"mimo-v2.5-pro","variant":"mimo-v2.5-pro","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1477,"score_low":1471.6,"score_high":1482.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"mimo-v2.5-pro","variantLabel":"mimo-v2.5-pro","score":1477,"secondary":{"rank":59,"rankSpread":[56,63],"ciPlus":5.4,"ciMinus":5.4,"ratingRaw":1476.9175289340246,"votes":20096,"modelKey":"mimo-v2.5-pro-webdev","modelOrganization":"Xiaomi","pricePerM":"$0.435 / $0.87","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:24e3159d841ef7a4","board":"arena_code","model":"kimi-k2.7-code","variant":"kimi-k2.7-code","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1473,"score_low":1463.6,"score_high":1482.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"kimi-k2.7-code","variantLabel":"kimi-k2.7-code","score":1473,"secondary":{"rank":60,"rankSpread":[56,65],"ciPlus":9.4,"ciMinus":9.4,"ratingRaw":1472.8739094118287,"votes":4992,"modelKey":"kimi-k2.7-code","modelOrganization":"Moonshot","pricePerM":"$0.6712 / $3.35","contextLength":262144},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:e98573cb34bd81ce","board":"arena_code","model":"deepseek-v4-pro","variant":"deepseek-v4-pro-high-preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1463,"score_low":1456.7,"score_high":1469.3,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"deepseek-v4-pro","variantLabel":"deepseek-v4-pro-high-preview","score":1463,"secondary":{"rank":63,"rankSpread":[59,67],"ciPlus":6.3,"ciMinus":6.3,"ratingRaw":1463.4472425933864,"votes":13223,"modelKey":"deepseek-v4-pro-ch1-thinking-webdev","modelOrganization":"DeepSeek","pricePerM":"$1.32 / $3.96","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:b7e5e5b8df1a65a0","board":"arena_code","model":"qwen3.6-plus","variant":"qwen3.6-plus","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1461,"score_low":1455.4,"score_high":1466.6,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"qwen3.6-plus","variantLabel":"qwen3.6-plus","score":1461,"secondary":{"rank":64,"rankSpread":[60,67],"ciPlus":5.6,"ciMinus":5.6,"ratingRaw":1460.5754178532447,"votes":18685,"modelKey":"qwen3.6-plus-v1-webdev","modelOrganization":"Alibaba","pricePerM":"$0.325 / $1.95","contextLength":1000000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:d0396ac7d382d35b","board":"arena_code","model":"gpt-5.5","variant":"gpt-5.5 (codex-harness)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"Codex","reasoning_effort":null,"service_mode":null,"score":1455,"score_low":1449.1,"score_high":1460.9,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gpt-5.5","variantLabel":"gpt-5.5 (codex-harness)","score":1455,"secondary":{"rank":65,"rankSpread":[62,69],"ciPlus":5.9,"ciMinus":5.9,"ratingRaw":1455.3189568528228,"votes":16045,"modelKey":"gpt-5.5-webdev","modelOrganization":"OpenAI","pricePerM":"$5 / $30","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:ccbd7f0883547751","board":"arena_code","model":"gemini-3.1-pro","variant":"gemini-3.1-pro-preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1446,"score_low":1441,"score_high":1451,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"gemini-3.1-pro","variantLabel":"gemini-3.1-pro-preview","score":1446,"secondary":{"rank":66,"rankSpread":[64,74],"ciPlus":5,"ciMinus":5,"ratingRaw":1446.2401469008362,"votes":24319,"modelKey":"gemini-3.1-pro-preview","modelOrganization":"Google","pricePerM":"$1 / $6","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:fec2568cdfba26c6","board":"arena_code","model":"deepseek-v4-pro","variant":"deepseek-v4-pro","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1446,"score_low":1439.7,"score_high":1452.3,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"deepseek-v4-pro","variantLabel":"deepseek-v4-pro","score":1446,"secondary":{"rank":67,"rankSpread":[64,76],"ciPlus":6.3,"ciMinus":6.3,"ratingRaw":1446.2181765174203,"votes":14032,"modelKey":"deepseek-v4-pro-ch1-webdev","modelOrganization":"DeepSeek","pricePerM":"$1.32 / $3.96","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:101c12e441513d87","board":"arena_code","model":"mimo-v2.5","variant":"mimo-v2.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1438,"score_low":1432,"score_high":1444,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"mimo-v2.5","variantLabel":"mimo-v2.5","score":1438,"secondary":{"rank":72,"rankSpread":[66,79],"ciPlus":6,"ciMinus":6,"ratingRaw":1437.6525376702152,"votes":15017,"modelKey":"mimo-v2.5-webdev","modelOrganization":"Xiaomi","pricePerM":"$0.14 / $0.28","contextLength":1050000},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:79ddd44617f3ec62","board":"arena_code","model":"kimi-k2.5","variant":"kimi-k2.5-thinking","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"thinking","service_mode":null,"score":1436,"score_low":1430.8,"score_high":1441.2,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"kimi-k2.5","variantLabel":"kimi-k2.5-thinking","score":1436,"secondary":{"rank":73,"rankSpread":[67,79],"ciPlus":5.2,"ciMinus":5.2,"ratingRaw":1435.8220231666578,"votes":20139,"modelKey":"kimi-k2.5-webdev","modelOrganization":"Moonshot","pricePerM":"$0.6 / $3","contextLength":null},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:1744764b0671d723","board":"arena_code","model":"glm-5","variant":"glm-5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1434,"score_low":1425.9,"score_high":1442.1,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"glm-5","variantLabel":"glm-5","score":1434,"secondary":{"rank":75,"rankSpread":[66,79],"ciPlus":8.1,"ciMinus":8.1,"ratingRaw":1434.004063616398,"votes":7505,"modelKey":"glm-5-webdev","modelOrganization":"Z.ai","pricePerM":"$1 / $3.2","contextLength":202800},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:f3628e90994fda44","board":"arena_code","model":"deepseek-v4-flash","variant":"deepseek-v4-flash-high-preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1430,"score_low":1423.1,"score_high":1436.9,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"deepseek-v4-flash","variantLabel":"deepseek-v4-flash-high-preview","score":1430,"secondary":{"rank":77,"rankSpread":[68,79],"ciPlus":6.9,"ciMinus":6.9,"ratingRaw":1430.270050604899,"votes":10152,"modelKey":"deepseek-v4-ch3-thinking-webdev","modelOrganization":"DeepSeek","pricePerM":"$0.3 / $1.2","contextLength":1048576},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:07b745ae2862d994","board":"arena_code","model":"inkling","variant":"inkling","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1412,"score_low":1405.6,"score_high":1418.4,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"inkling","variantLabel":"inkling","score":1412,"secondary":{"rank":80,"rankSpread":[77,86],"ciPlus":6.4,"ciMinus":6.4,"ratingRaw":1411.9713399189854,"votes":13499,"modelKey":"inkling-webdev","modelOrganization":"Thinky","pricePerM":"$1 / $4.05","contextLength":524288},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:5e9cc8edc5997bcf","board":"arena_code","model":"inkling-small","variant":"Inkling Small","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1408,"score_low":1399,"score_high":1417,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"inkling-small","variantLabel":"Inkling Small","score":1408,"secondary":{"rank":81,"rankSpread":[77,91],"ciPlus":9,"ciMinus":9,"ratingRaw":1408.2923972599388,"votes":5420,"modelKey":"inkling-small","modelOrganization":"Thinky","pricePerM":"$0.45 / $1.2","contextLength":524288},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:cbe1c293dbc12cfd","board":"arena_code","model":"kimi-k2.5","variant":"kimi-k2.5-instant","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1404,"score_low":1392,"score_high":1416,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"kimi-k2.5","variantLabel":"kimi-k2.5-instant","score":1404,"secondary":{"rank":83,"rankSpread":[77,93],"ciPlus":12,"ciMinus":12,"ratingRaw":1403.648568655793,"votes":3266,"modelKey":"kimi-k2.5-instant","modelOrganization":"Moonshot","pricePerM":"$0.45 / $2.25","contextLength":262144},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:891df129a594a41e","board":"arena_code","model":"minimax-m2.7","variant":"minimax-m2.7","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1397,"score_low":1391.2,"score_high":1402.8,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"minimax-m2.7","variantLabel":"minimax-m2.7","score":1397,"secondary":{"rank":86,"rankSpread":[78,93],"ciPlus":5.8,"ciMinus":5.8,"ratingRaw":1397.0863812400469,"votes":16079,"modelKey":"deep-octo","modelOrganization":"MiniMax","pricePerM":"$0.3 / $1.2","contextLength":204800},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:28d1b5355f79eb7b","board":"arena_code","model":"minimax-m2.5","variant":"minimax-m2.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1386,"score_low":1377.7,"score_high":1394.3,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"minimax-m2.5","variantLabel":"minimax-m2.5","score":1386,"secondary":{"rank":92,"rankSpread":[81,98],"ciPlus":8.3,"ciMinus":8.3,"ratingRaw":1386.2682419835223,"votes":7628,"modelKey":"minimax-m2.5","modelOrganization":"MiniMax","pricePerM":"$0.27 / $1.08","contextLength":204800},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_code:daa8896f14bb8dc8","board":"arena_code","model":"mistral-medium-3.5","variant":"mistral-medium-3.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":1263,"score_low":1248.2,"score_high":1277.8,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":null,"source":"https://arena.ai/leaderboard/code","checked_at":"2026-10-01","archive":"scores-code-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_code","model":"mistral-medium-3.5","variantLabel":"mistral-medium-3.5","score":1263,"secondary":{"rank":121,"rankSpread":[118,129],"ciPlus":14.8,"ciMinus":14.8,"ratingRaw":1263.107483617966,"votes":2318,"modelKey":"mistral-medium-3.5-webdev","modelOrganization":"Mistral","pricePerM":"$0.75 / $3.75","contextLength":262144},"source":"https://arena.ai/leaderboard/code","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:d1cdd70383ae2039","board":"arena_agent_mode","model":"claude-fable-5.1","variant":"Claude Fable 5.1 (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":14.55,"score_low":12.63,"score_high":16.47,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":4.44,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-fable-5.1","variantLabel":"Claude Fable 5.1 (Max)","score":14.55,"secondary":{"rank":1,"rankSpread":[1,4],"ciPlus":1.92,"ciMinus":1.92,"netImprovementRaw":0.1455297025193051,"pipelines":5,"medianCostPerTaskUsd":4.44,"medianCostPerTaskUsdRaw":4.436707999999999,"sessions":14636,"contenderName":"contenders/claude-fable-5.1-max-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:b1592396c4d6d498","board":"arena_agent_mode","model":"claude-opus-5.5","variant":"Claude Opus 5.5 (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":13.78,"score_low":11.54,"score_high":16.02,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":1.57,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-opus-5.5","variantLabel":"Claude Opus 5.5 (High)","score":13.78,"secondary":{"rank":2,"rankSpread":[1,4],"ciPlus":2.24,"ciMinus":2.24,"netImprovementRaw":0.13784181616281832,"pipelines":5,"medianCostPerTaskUsd":1.57,"medianCostPerTaskUsdRaw":1.5690757,"sessions":5013,"contenderName":"contenders/claude-opus-5.5-high-vertex-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:704fea58650c673b","board":"arena_agent_mode","model":"gpt-6-astra","variant":"GPT 6 Astra (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":12.18,"score_low":9.9,"score_high":14.459999999999999,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":2.9,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-6-astra","variantLabel":"GPT 6 Astra (Max)","score":12.18,"secondary":{"rank":3,"rankSpread":[1,7],"ciPlus":2.28,"ciMinus":2.28,"netImprovementRaw":0.12175489332353712,"pipelines":5,"medianCostPerTaskUsd":2.9,"medianCostPerTaskUsdRaw":2.9008555,"sessions":11077,"contenderName":"contenders/gpt-6-astra-max-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:c3f5ba38fbb30407","board":"arena_agent_mode","model":"gpt-6-sol","variant":"GPT 6 Sol (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":10.65,"score_low":8.08,"score_high":13.22,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.91,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-6-sol","variantLabel":"GPT 6 Sol (Max)","score":10.65,"secondary":{"rank":4,"rankSpread":[1,10],"ciPlus":2.57,"ciMinus":2.57,"netImprovementRaw":0.10645668688061616,"pipelines":5,"medianCostPerTaskUsd":0.91,"medianCostPerTaskUsdRaw":0.9061006000000003,"sessions":5031,"contenderName":"contenders/gpt-6-sol-max-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:63ab115851e0de58","board":"arena_agent_mode","model":"claude-opus-5","variant":"Claude Opus 5 (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":8.76,"score_low":7.26,"score_high":10.26,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":2.5,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-opus-5","variantLabel":"Claude Opus 5 (High)","score":8.76,"secondary":{"rank":5,"rankSpread":[3,10],"ciPlus":1.5,"ciMinus":1.5,"netImprovementRaw":0.08760655303538206,"pipelines":5,"medianCostPerTaskUsd":2.5,"medianCostPerTaskUsdRaw":2.497321875,"sessions":26501,"contenderName":"contenders/claude-opus-5-high-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:7b426212aba5971e","board":"arena_agent_mode","model":"claude-opus-5","variant":"Claude Opus 5 (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":8.55,"score_low":6.98,"score_high":10.120000000000001,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":3.59,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-opus-5","variantLabel":"Claude Opus 5 (Max)","score":8.55,"secondary":{"rank":6,"rankSpread":[3,10],"ciPlus":1.57,"ciMinus":1.57,"netImprovementRaw":0.085451073507389,"pipelines":5,"medianCostPerTaskUsd":3.59,"medianCostPerTaskUsdRaw":3.5855317500000004,"sessions":21714,"contenderName":"contenders/claude-opus-5-max-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:3d3ae04ef0c99602","board":"arena_agent_mode","model":"claude-fable-5","variant":"Claude Fable 5 (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":8.37,"score_low":7.069999999999999,"score_high":9.67,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":2.02,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-fable-5","variantLabel":"Claude Fable 5 (High)","score":8.37,"secondary":{"rank":7,"rankSpread":[4,10],"ciPlus":1.3,"ciMinus":1.3,"netImprovementRaw":0.08366053240098753,"pipelines":5,"medianCostPerTaskUsd":2.02,"medianCostPerTaskUsdRaw":2.0244934999999997,"sessions":39998,"contenderName":"contenders/claude-fable-5-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:5c55fe7f551fe2c6","board":"arena_agent_mode","model":"gpt-5.6-sol","variant":"GPT 5.6 Sol (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":7.05,"score_low":5.67,"score_high":8.43,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":1.05,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-5.6-sol","variantLabel":"GPT 5.6 Sol (xHigh)","score":7.05,"secondary":{"rank":9,"rankSpread":[4,11],"ciPlus":1.38,"ciMinus":1.38,"netImprovementRaw":0.07054582147602956,"pipelines":5,"medianCostPerTaskUsd":1.05,"medianCostPerTaskUsdRaw":1.05124595,"sessions":33916,"contenderName":"contenders/gpt-5.6-sol-xhigh-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:263865305f7ecdb7","board":"arena_agent_mode","model":"claude-opus-4.8","variant":"Claude Opus 4.8 (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":6.92,"score_low":5.54,"score_high":8.3,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":1.36,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-opus-4.8","variantLabel":"Claude Opus 4.8 (High)","score":6.92,"secondary":{"rank":10,"rankSpread":[4,11],"ciPlus":1.38,"ciMinus":1.38,"netImprovementRaw":0.06923697104299241,"pipelines":5,"medianCostPerTaskUsd":1.36,"medianCostPerTaskUsdRaw":1.3621281666666665,"sessions":40210,"contenderName":"contenders/claude-opus-4-8-thinking-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:5729b5a20a335dfc","board":"arena_agent_mode","model":"claude-sonnet-5","variant":"Claude Sonnet 5 (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":4.81,"score_low":3.09,"score_high":6.529999999999999,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.9,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"claude-sonnet-5","variantLabel":"Claude Sonnet 5 (High)","score":4.81,"secondary":{"rank":11,"rankSpread":[8,22],"ciPlus":1.72,"ciMinus":1.72,"netImprovementRaw":0.048060893931133726,"pipelines":5,"medianCostPerTaskUsd":0.9,"medianCostPerTaskUsdRaw":0.897684725,"sessions":31480,"contenderName":"contenders/claude-sonnet-5-agent","modelOrganization":"Anthropic"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:edee122a50e59a89","board":"arena_agent_mode","model":"hy4-preview","variant":"Hy4 preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":4.24,"score_low":3.39,"score_high":5.09,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.29,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"hy4-preview","variantLabel":"Hy4 preview","score":4.24,"secondary":{"rank":12,"rankSpread":[10,20],"ciPlus":0.85,"ciMinus":0.85,"netImprovementRaw":0.04244555997749705,"pipelines":5,"medianCostPerTaskUsd":0.29,"medianCostPerTaskUsdRaw":0.29312866133333326,"sessions":53732,"contenderName":"contenders/phoenix-qcs2-agent","modelOrganization":"Tencent"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:0d8b2c85e74d262f","board":"arena_agent_mode","model":"muse-spark-1.3","variant":"Muse Spark 1.3 (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":4.19,"score_low":3.6000000000000005,"score_high":4.78,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.41,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"muse-spark-1.3","variantLabel":"Muse Spark 1.3 (Max)","score":4.19,"secondary":{"rank":13,"rankSpread":[10,19],"ciPlus":0.59,"ciMinus":0.59,"netImprovementRaw":0.041918390470629244,"pipelines":5,"medianCostPerTaskUsd":0.41,"medianCostPerTaskUsdRaw":0.40948609999999996,"sessions":61422,"contenderName":"contenders/muse-spark-1-3-max-agent","modelOrganization":"Meta"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:605f40536c85bbb4","board":"arena_agent_mode","model":"gpt-5.5","variant":"GPT 5.5 (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":4.12,"score_low":3.0700000000000003,"score_high":5.17,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":1.22,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-5.5","variantLabel":"GPT 5.5 (xHigh)","score":4.12,"secondary":{"rank":14,"rankSpread":[10,22],"ciPlus":1.05,"ciMinus":1.05,"netImprovementRaw":0.041160605697829145,"pipelines":5,"medianCostPerTaskUsd":1.22,"medianCostPerTaskUsdRaw":1.221792,"sessions":53913,"contenderName":"contenders/gpt-5.5-xhigh-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:c8965e5dd2526f05","board":"arena_agent_mode","model":"deepseek-v4.1-flash","variant":"Deepseek V4.1 Flash (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":4.01,"score_low":3.51,"score_high":4.51,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.09,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"deepseek-v4.1-flash","variantLabel":"Deepseek V4.1 Flash (Max)","score":4.01,"secondary":{"rank":15,"rankSpread":[11,19],"ciPlus":0.5,"ciMinus":0.5,"netImprovementRaw":0.04005070128566882,"pipelines":5,"medianCostPerTaskUsd":0.09,"medianCostPerTaskUsdRaw":0.093749272,"sessions":95889,"contenderName":"contenders/deepseek-v4.1-flash-max-20260910-agent","modelOrganization":"DeepSeek"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:1307216843d89276","board":"arena_agent_mode","model":"kimi-k3","variant":"Kimi K3 (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":3.98,"score_low":3.45,"score_high":4.51,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.75,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"kimi-k3","variantLabel":"Kimi K3 (Max)","score":3.98,"secondary":{"rank":16,"rankSpread":[11,20],"ciPlus":0.53,"ciMinus":0.53,"netImprovementRaw":0.03983378687688044,"pipelines":5,"medianCostPerTaskUsd":0.75,"medianCostPerTaskUsdRaw":0.7508934,"sessions":129908,"contenderName":"contenders/kivine-wxzc-agent","modelOrganization":"Moonshot"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:0684f9cc2a945bdb","board":"arena_agent_mode","model":"grok-4.7","variant":"Grok 4.7 (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":3.64,"score_low":2.0700000000000003,"score_high":5.21,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":1.47,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"grok-4.7","variantLabel":"Grok 4.7 (xHigh)","score":3.64,"secondary":{"rank":17,"rankSpread":[10,27],"ciPlus":1.57,"ciMinus":1.57,"netImprovementRaw":0.03641197380079611,"pipelines":5,"medianCostPerTaskUsd":1.47,"medianCostPerTaskUsdRaw":1.4728959999999998,"sessions":10109,"contenderName":"contenders/grok-4.7-xhigh-public-agent","modelOrganization":"SpaceXAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:f4f60f98940c49ee","board":"arena_agent_mode","model":"glm-5.2","variant":"GLM 5.2 (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":3.64,"score_low":2.9000000000000004,"score_high":4.38,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.76,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"glm-5.2","variantLabel":"GLM 5.2 (Max)","score":3.64,"secondary":{"rank":18,"rankSpread":[11,23],"ciPlus":0.74,"ciMinus":0.74,"netImprovementRaw":0.0363506120697182,"pipelines":5,"medianCostPerTaskUsd":0.76,"medianCostPerTaskUsdRaw":0.75712883,"sessions":83747,"contenderName":"contenders/glm-5.2-agent","modelOrganization":"Z.ai"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:60a00778bc1492a5","board":"arena_agent_mode","model":"gemini-3.8-flash","variant":"Gemini 3.8 Flash (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":2.96,"score_low":2.14,"score_high":3.78,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.58,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gemini-3.8-flash","variantLabel":"Gemini 3.8 Flash (High)","score":2.96,"secondary":{"rank":19,"rankSpread":[11,24],"ciPlus":0.82,"ciMinus":0.82,"netImprovementRaw":0.029563106907863203,"pipelines":5,"medianCostPerTaskUsd":0.58,"medianCostPerTaskUsdRaw":0.5811470812499999,"sessions":32890,"contenderName":"contenders/skimaki-kwc6-agent","modelOrganization":"Google"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:9ce366a4c245bae4","board":"arena_agent_mode","model":"qwen3.8-max","variant":"Qwen3.8 Max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":2.8,"score_low":2.15,"score_high":3.4499999999999997,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.44,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"qwen3.8-max","variantLabel":"Qwen3.8 Max","score":2.8,"secondary":{"rank":20,"rankSpread":[13,24],"ciPlus":0.65,"ciMinus":0.65,"netImprovementRaw":0.02804376354442393,"pipelines":5,"medianCostPerTaskUsd":0.44,"medianCostPerTaskUsdRaw":0.43885050000000003,"sessions":49157,"contenderName":"contenders/paisley-n9x0","modelOrganization":"Alibaba"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:64885804c8dddecd","board":"arena_agent_mode","model":"glm-5.3","variant":"GLM 5.3 (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":2.57,"score_low":1.8499999999999999,"score_high":3.29,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.64,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"glm-5.3","variantLabel":"GLM 5.3 (Max)","score":2.57,"secondary":{"rank":21,"rankSpread":[15,27],"ciPlus":0.72,"ciMinus":0.72,"netImprovementRaw":0.02574044983167535,"pipelines":5,"medianCostPerTaskUsd":0.64,"medianCostPerTaskUsdRaw":0.64470332,"sessions":80704,"contenderName":"contenders/glm-5.3-agent","modelOrganization":"Z.ai"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:04bd9ba65c143398","board":"arena_agent_mode","model":"gpt-6-luna","variant":"GPT 6 Luna (Max)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":1.68,"score_low":0.27,"score_high":3.09,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.06,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-6-luna","variantLabel":"GPT 6 Luna (Max)","score":1.68,"secondary":{"rank":22,"rankSpread":[15,28],"ciPlus":1.41,"ciMinus":1.41,"netImprovementRaw":0.01680924574841998,"pipelines":5,"medianCostPerTaskUsd":0.06,"medianCostPerTaskUsdRaw":0.06303544,"sessions":12076,"contenderName":"contenders/gpt-6-luna-max-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:3081d8b084b897bf","board":"arena_agent_mode","model":"deepseek-v4-pro","variant":"DeepSeek V4 Pro (High) (0813)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":1.29,"score_low":-0.4099999999999999,"score_high":2.99,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.38,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"deepseek-v4-pro","variantLabel":"DeepSeek V4 Pro (High) (0813)","score":1.29,"secondary":{"rank":23,"rankSpread":[17,31],"ciPlus":1.7,"ciMinus":1.7,"netImprovementRaw":0.012890258994477245,"pipelines":5,"medianCostPerTaskUsd":0.38,"medianCostPerTaskUsdRaw":0.3779027875,"sessions":6743,"contenderName":"contenders/deepseek-v4-pro-high-fireworks","modelOrganization":"DeepSeek"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:09fac21036f59279","board":"arena_agent_mode","model":"grok-4.6","variant":"Grok 4.6 (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":1.16,"score_low":0.1399999999999999,"score_high":2.1799999999999997,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.99,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"grok-4.6","variantLabel":"Grok 4.6 (xHigh)","score":1.16,"secondary":{"rank":24,"rankSpread":[18,29],"ciPlus":1.02,"ciMinus":1.02,"netImprovementRaw":0.01156080975145234,"pipelines":5,"medianCostPerTaskUsd":0.99,"medianCostPerTaskUsdRaw":0.988828,"sessions":25563,"contenderName":"contenders/grok-4.6-xhigh-public-agent","modelOrganization":"SpaceXAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:f89b60d1061b77e1","board":"arena_agent_mode","model":"gpt-5.5","variant":"GPT 5.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1.13,"score_low":0.15999999999999992,"score_high":2.0999999999999996,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.61,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-5.5","variantLabel":"GPT 5.5","score":1.13,"secondary":{"rank":25,"rankSpread":[20,29],"ciPlus":0.97,"ciMinus":0.97,"netImprovementRaw":0.011277335271279987,"pipelines":5,"medianCostPerTaskUsd":0.61,"medianCostPerTaskUsdRaw":0.610801,"sessions":82103,"contenderName":"contenders/gpt-5.5-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:069dfce97d9ef41d","board":"arena_agent_mode","model":"grok-4.5","variant":"Grok 4.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":1.11,"score_low":0.1200000000000001,"score_high":2.1,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.44,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"grok-4.5","variantLabel":"Grok 4.5","score":1.11,"secondary":{"rank":26,"rankSpread":[20,29],"ciPlus":0.99,"ciMinus":0.99,"netImprovementRaw":0.011084195756121486,"pipelines":5,"medianCostPerTaskUsd":0.44,"medianCostPerTaskUsdRaw":0.43587000000000004,"sessions":41479,"contenderName":"contenders/grok-4.5-agent","modelOrganization":"SpaceXAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:96e32efd63046574","board":"arena_agent_mode","model":"gpt-5.6-terra","variant":"GPT 5.6 Terra (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":0.83,"score_low":-0.42000000000000004,"score_high":2.08,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.43,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-5.6-terra","variantLabel":"GPT 5.6 Terra (xHigh)","score":0.83,"secondary":{"rank":28,"rankSpread":[20,31],"ciPlus":1.25,"ciMinus":1.25,"netImprovementRaw":0.008294417032596956,"pipelines":5,"medianCostPerTaskUsd":0.43,"medianCostPerTaskUsdRaw":0.42624249999999997,"sessions":22046,"contenderName":"contenders/gpt-5.6-terra-xhigh-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:bd0ff69aca5acca9","board":"arena_agent_mode","model":"glm-5.3-flash","variant":"GLM 5.3 Flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-0.28,"score_low":-0.76,"score_high":0.19999999999999996,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.1,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"glm-5.3-flash","variantLabel":"GLM 5.3 Flash","score":-0.28,"secondary":{"rank":29,"rankSpread":[23,31],"ciPlus":0.48,"ciMinus":0.48,"netImprovementRaw":-0.0028286401774680115,"pipelines":5,"medianCostPerTaskUsd":0.1,"medianCostPerTaskUsdRaw":0.100495625,"sessions":76297,"contenderName":"contenders/glm-5.3-flash-agent","modelOrganization":"Z.ai"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:8455f3eb58f47b3d","board":"arena_agent_mode","model":"gpt-5.6-luna","variant":"GPT 5.6 Luna (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":-1.09,"score_low":-1.79,"score_high":-0.3900000000000001,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.07,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gpt-5.6-luna","variantLabel":"GPT 5.6 Luna (xHigh)","score":-1.09,"secondary":{"rank":31,"rankSpread":[27,33],"ciPlus":0.7,"ciMinus":0.7,"netImprovementRaw":-0.010900954420771746,"pipelines":5,"medianCostPerTaskUsd":0.07,"medianCostPerTaskUsdRaw":0.07296761,"sessions":39830,"contenderName":"contenders/gpt-5.6-luna-xhigh-agent","modelOrganization":"OpenAI"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:22652d46f50fe772","board":"arena_agent_mode","model":"qwen3.8-27b","variant":"Qwen 3.8 27B","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-1.61,"score_low":-2.21,"score_high":-1.0100000000000002,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.31,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"qwen3.8-27b","variantLabel":"Qwen 3.8 27B","score":-1.61,"secondary":{"rank":32,"rankSpread":[30,33],"ciPlus":0.6,"ciMinus":0.6,"netImprovementRaw":-0.016128490732088732,"pipelines":5,"medianCostPerTaskUsd":0.31,"medianCostPerTaskUsdRaw":0.3128092,"sessions":53985,"contenderName":"contenders/qwen3.8-27b-code-agent","modelOrganization":"Alibaba"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:47cad90e68374509","board":"arena_agent_mode","model":"gemini-3.7-flash","variant":"Gemini 3.7 Flash (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":-1.64,"score_low":-2.4,"score_high":-0.8799999999999999,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.38,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gemini-3.7-flash","variantLabel":"Gemini 3.7 Flash (High)","score":-1.64,"secondary":{"rank":33,"rankSpread":[30,34],"ciPlus":0.76,"ciMinus":0.76,"netImprovementRaw":-0.016418883599838806,"pipelines":5,"medianCostPerTaskUsd":0.38,"medianCostPerTaskUsdRaw":0.3759328125,"sessions":51614,"contenderName":"contenders/andwise-evfd-agent","modelOrganization":"Google"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:84c2affbfb9336da","board":"arena_agent_mode","model":"muse-spark-1.2","variant":"Muse Spark 1.2 (xHigh)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"xhigh","service_mode":null,"score":-2.96,"score_low":-3.65,"score_high":-2.27,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.4,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"muse-spark-1.2","variantLabel":"Muse Spark 1.2 (xHigh)","score":-2.96,"secondary":{"rank":34,"rankSpread":[33,34],"ciPlus":0.69,"ciMinus":0.69,"netImprovementRaw":-0.029574610776391144,"pipelines":5,"medianCostPerTaskUsd":0.4,"medianCostPerTaskUsdRaw":0.39728209999999997,"sessions":46284,"contenderName":"contenders/muse-spark-1.2-xhigh-agent","modelOrganization":"Meta"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:15166add541e68d5","board":"arena_agent_mode","model":"qwen3.7-max","variant":"Qwen3.7 Max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"max","service_mode":null,"score":-5.2,"score_low":-6.140000000000001,"score_high":-4.26,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.19,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"qwen3.7-max","variantLabel":"Qwen3.7 Max","score":-5.2,"secondary":{"rank":36,"rankSpread":[35,39],"ciPlus":0.94,"ciMinus":0.94,"netImprovementRaw":-0.05202787044973144,"pipelines":5,"medianCostPerTaskUsd":0.19,"medianCostPerTaskUsdRaw":0.1949871825,"sessions":39957,"contenderName":"contenders/qwen3.7-max-agent","modelOrganization":"Alibaba"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:3a37aca313fcd27d","board":"arena_agent_mode","model":"hy3","variant":"Hy3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-5.7,"score_low":-6.93,"score_high":-4.470000000000001,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.04,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"hy3","variantLabel":"Hy3","score":-5.7,"secondary":{"rank":37,"rankSpread":[35,42],"ciPlus":1.23,"ciMinus":1.23,"netImprovementRaw":-0.057044187117040346,"pipelines":5,"medianCostPerTaskUsd":0.04,"medianCostPerTaskUsdRaw":0.044244222,"sessions":29011,"contenderName":"contenders/hy3-tencent-cloud-agent","modelOrganization":"Tencent"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:a16fe8d7c7b78883","board":"arena_agent_mode","model":"minimax-m3","variant":"Minimax M3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-6.71,"score_low":-7.6,"score_high":-5.82,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.08,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"minimax-m3","variantLabel":"Minimax M3","score":-6.71,"secondary":{"rank":38,"rankSpread":[36,42],"ciPlus":0.89,"ciMinus":0.89,"netImprovementRaw":-0.06710541285201452,"pipelines":5,"medianCostPerTaskUsd":0.08,"medianCostPerTaskUsdRaw":0.08431896,"sessions":40771,"contenderName":"contenders/minimax-m3-first-party","modelOrganization":"MiniMax"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:c6879efca1b5dffc","board":"arena_agent_mode","model":"mimo-v2.5-pro","variant":"Mimo V2.5 Pro","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-7.23,"score_low":-8.14,"score_high":-6.32,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.04,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"mimo-v2.5-pro","variantLabel":"Mimo V2.5 Pro","score":-7.23,"secondary":{"rank":39,"rankSpread":[37,42],"ciPlus":0.91,"ciMinus":0.91,"netImprovementRaw":-0.07228507779222011,"pipelines":5,"medianCostPerTaskUsd":0.04,"medianCostPerTaskUsdRaw":0.038483031,"sessions":44405,"contenderName":"contenders/mimo-v2.5-pro-agent","modelOrganization":"Xiaomi"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:9cf4eb892a450683","board":"arena_agent_mode","model":"qwen3.7-plus","variant":"Qwen3.7 Plus","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-7.28,"score_low":-8.75,"score_high":-5.8100000000000005,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.09,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"qwen3.7-plus","variantLabel":"Qwen3.7 Plus","score":-7.28,"secondary":{"rank":40,"rankSpread":[36,43],"ciPlus":1.47,"ciMinus":1.47,"netImprovementRaw":-0.07278309668346637,"pipelines":5,"medianCostPerTaskUsd":0.09,"medianCostPerTaskUsdRaw":0.092556064,"sessions":21406,"contenderName":"contenders/qwen3.7-plus-agent","modelOrganization":"Alibaba"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:46d0083fa34f1126","board":"arena_agent_mode","model":"gemini-3.1-pro","variant":"Gemini 3.1 Pro Preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-7.35,"score_low":-8.33,"score_high":-6.369999999999999,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.29,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gemini-3.1-pro","variantLabel":"Gemini 3.1 Pro Preview","score":-7.35,"secondary":{"rank":41,"rankSpread":[37,42],"ciPlus":0.98,"ciMinus":0.98,"netImprovementRaw":-0.07349666328844258,"pipelines":5,"medianCostPerTaskUsd":0.29,"medianCostPerTaskUsdRaw":0.2931636,"sessions":86482,"contenderName":"contenders/gemini-3.1-pro-preview-agent","modelOrganization":"Google"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:ee3d29f57f6b6e33","board":"arena_agent_mode","model":"gemini-3.6-flash","variant":"Gemini 3.6 Flash (High)","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"high","service_mode":null,"score":-7.67,"score_low":-8.75,"score_high":-6.59,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.3,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"gemini-3.6-flash","variantLabel":"Gemini 3.6 Flash (High)","score":-7.67,"secondary":{"rank":42,"rankSpread":[37,43],"ciPlus":1.08,"ciMinus":1.08,"netImprovementRaw":-0.07674899660899959,"pipelines":5,"medianCostPerTaskUsd":0.3,"medianCostPerTaskUsdRaw":0.295381275,"sessions":22568,"contenderName":"contenders/gemini-3.6-flash-agent","modelOrganization":"Google"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:03dbaa1180987af9","board":"arena_agent_mode","model":"inkling-small","variant":"Inkling Small","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-9.8,"score_low":-11.22,"score_high":-8.38,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.1,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"inkling-small","variantLabel":"Inkling Small","score":-9.8,"secondary":{"rank":43,"rankSpread":[41,45],"ciPlus":1.42,"ciMinus":1.42,"netImprovementRaw":-0.09797549493301351,"pipelines":5,"medianCostPerTaskUsd":0.1,"medianCostPerTaskUsdRaw":0.0992085,"sessions":13028,"contenderName":"contenders/inkling-small","modelOrganization":"Thinky"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:8967f33635820f8b","board":"arena_agent_mode","model":"inkling","variant":"Inkling","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":null,"service_mode":null,"score":-10.9,"score_low":-11.96,"score_high":-9.84,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.25,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"inkling","variantLabel":"Inkling","score":-10.9,"secondary":{"rank":44,"rankSpread":[43,45],"ciPlus":1.06,"ciMinus":1.06,"netImprovementRaw":-0.10904642553845262,"pipelines":5,"medianCostPerTaskUsd":0.25,"medianCostPerTaskUsdRaw":0.24988325,"sessions":42596,"contenderName":"contenders/inkling-agent","modelOrganization":"Thinky"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"arena_agent_mode:9c4432eeedad5bad","board":"arena_agent_mode","model":"mistral-medium-3.5","variant":"Mistral Medium 3.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":null,"reasoning_effort":"medium","service_mode":null,"score":-11.74,"score_low":-13.24,"score_high":-10.24,"mean_cost_usd_per_task":null,"median_cost_usd_per_task":0.67,"source":"https://arena.ai/leaderboard/agent","checked_at":"2026-10-01","archive":"scores-agent-arena-round2-2026-10-01.json","raw_record":{"boardId":"arena_agent_mode","model":"mistral-medium-3.5","variantLabel":"Mistral Medium 3.5","score":-11.74,"secondary":{"rank":45,"rankSpread":[43,45],"ciPlus":1.5,"ciMinus":1.5,"netImprovementRaw":-0.11737227392552999,"pipelines":5,"medianCostPerTaskUsd":0.67,"medianCostPerTaskUsdRaw":0.670423,"sessions":10797,"contenderName":"contenders/mistral-medium-3.5-v2-agent","modelOrganization":"Mistral"},"source":"https://arena.ai/leaderboard/agent","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:1f49b4d125f7e8d5","board":"open_design_arena","model":"gpt-6.1-sol","variant":"GPT-6.1 Sol","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":90.1,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.38203,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"gpt-6.1-sol","variantLabel":"GPT-6.1 Sol","score":90.1,"secondary":{"rank":1,"rankBasis":"average_score_desc","officialName":"GPT-6.1 Sol","maker":"OpenAI","icon":"/agents/openai.svg","color":"var(--data-green)","pdr":90,"d0":null,"d1":28,"d2":62.2,"ttft":13.6,"time":372.8,"cacheHit":83,"inputTokens":389067.4,"outputTokens":23310.5,"cost":0.38203,"calls":8.7,"scenarios":[{"id":"webapp","n":7,"score":93.8,"ci95":4.3,"pdr":100,"d1":28.8,"d2":65.1,"ttft":12.2,"time":406.2,"cacheHit":83.3,"inputTokens":480316.9,"outputTokens":31661.7,"cost":0.490331,"calls":11.4},{"id":"mobile","n":9,"score":78.4,"ci95":19.9,"pdr":66.7,"d1":26.5,"d2":51.9,"ttft":15.9,"time":384.1,"cacheHit":85.4,"inputTokens":470078.6,"outputTokens":22244.4,"cost":0.383211,"calls":11.6},{"id":"desktop","n":7,"score":93,"ci95":5.1,"pdr":100,"d1":28.4,"d2":64.8,"ttft":12.4,"time":357.2,"cacheHit":80.2,"inputTokens":265959.1,"outputTokens":21450.6,"cost":0.340023,"calls":4.9},{"id":"dashboard","n":8,"score":95.6,"ci95":2.9,"pdr":100,"d1":29.7,"d2":65.9,"ttft":13.3,"time":363.3,"cacheHit":80.5,"inputTokens":277709.6,"outputTokens":23439.4,"cost":0.364905,"calls":5.3},{"id":"landing","n":9,"score":89.8,"ci95":5.6,"pdr":88.9,"d1":26.6,"d2":63.3,"ttft":13.6,"time":353,"cacheHit":84.9,"inputTokens":445498.9,"outputTokens":19419.7,"cost":0.349178,"calls":10.2}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:63b060c2709bd850","board":"open_design_arena","model":"gpt-6-sol","variant":"GPT-6 Sol","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":89.9,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.296641,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"gpt-6-sol","variantLabel":"GPT-6 Sol","score":89.9,"secondary":{"rank":2,"rankBasis":"average_score_desc","officialName":"GPT-6 Sol","maker":"OpenAI","icon":"/agents/openai.svg","color":"var(--data-coral)","pdr":85.7,"d0":null,"d1":27.7,"d2":62.2,"ttft":32.3,"time":284.9,"cacheHit":81.5,"inputTokens":344403.5,"outputTokens":13705.6,"cost":0.296641,"calls":9.4,"scenarios":[{"id":"webapp","n":8,"score":90.5,"ci95":7.6,"pdr":75,"d1":27.5,"d2":63,"ttft":27.4,"time":307.4,"cacheHit":80.2,"inputTokens":325050.6,"outputTokens":13930.9,"cost":0.304231,"calls":10.1},{"id":"mobile","n":9,"score":87.1,"ci95":5.9,"pdr":77.8,"d1":28.8,"d2":58.3,"ttft":44.3,"time":318.1,"cacheHit":80.5,"inputTokens":297154.2,"outputTokens":13185.1,"cost":0.288846,"calls":9},{"id":"desktop","n":8,"score":90.4,"ci95":2.6,"pdr":100,"d1":26.5,"d2":63.9,"ttft":36.1,"time":251.5,"cacheHit":80.3,"inputTokens":210245.6,"outputTokens":13831,"cost":0.251772,"calls":6.4},{"id":"dashboard","n":8,"score":93.4,"ci95":3.8,"pdr":100,"d1":27.8,"d2":65.6,"ttft":22.3,"time":195.5,"cacheHit":81,"inputTokens":256252.6,"outputTokens":14545.1,"cost":0.280613,"calls":6.7},{"id":"landing","n":9,"score":88.6,"ci95":4.7,"pdr":77.8,"d1":27.9,"d2":60.7,"ttft":31.4,"time":351.9,"cacheHit":85.7,"inputTokens":633314.6,"outputTokens":13035.9,"cost":0.357744,"calls":14.6}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:5d69f844da6dc80d","board":"open_design_arena","model":"claude-sonnet-5.5","variant":"Claude Sonnet 5.5","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":86.8,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.573814,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"claude-sonnet-5.5","variantLabel":"Claude Sonnet 5.5","score":86.8,"secondary":{"rank":3,"rankBasis":"average_score_desc","officialName":"Claude Sonnet 5.5","maker":"Anthropic","icon":"/model-icons/claude-brand.svg","color":"var(--data-teal)","pdr":77.3,"d0":null,"d1":27.3,"d2":59.5,"ttft":48.2,"time":244.8,"cacheHit":49.8,"inputTokens":288407,"outputTokens":31318.3,"cost":0.573814,"calls":2.6,"scenarios":[{"id":"webapp","n":9,"score":86.4,"ci95":3.7,"pdr":77.8,"d1":27.3,"d2":59.1,"ttft":40.6,"time":281,"cacheHit":45.6,"inputTokens":287653.1,"outputTokens":41545.1,"cost":0.672946,"calls":2.4},{"id":"mobile","n":9,"score":86.4,"ci95":5.2,"pdr":77.8,"d1":29.6,"d2":56.8,"ttft":54,"time":278.1,"cacheHit":67.1,"inputTokens":433321.8,"outputTokens":34640.4,"cost":0.621203,"calls":4.3},{"id":"desktop","n":9,"score":91,"ci95":4.1,"pdr":100,"d1":28.8,"d2":62.2,"ttft":38.5,"time":254.2,"cacheHit":49.6,"inputTokens":302766.7,"outputTokens":36293.2,"cost":0.634123,"calls":2.9},{"id":"dashboard","n":9,"score":86.9,"ci95":8,"pdr":77.8,"d1":23.1,"d2":63.8,"ttft":52.4,"time":227.7,"cacheHit":44.3,"inputTokens":209492.1,"outputTokens":23862.8,"cost":0.484012,"calls":1.7},{"id":"landing","n":8,"score":82.9,"ci95":8.2,"pdr":50,"d1":27.8,"d2":55.1,"ttft":55.7,"time":182.8,"cacheHit":42.4,"inputTokens":208801.6,"outputTokens":20249.8,"cost":0.456788,"calls":1.4}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:af24dbf7d52582a5","board":"open_design_arena","model":"mimo-v2.6-flash","variant":"MiMo 2.6 Flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":83.5,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.02894,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"mimo-v2.6-flash","variantLabel":"MiMo 2.6 Flash","score":83.5,"secondary":{"rank":4,"rankBasis":"average_score_desc","officialName":"MiMo 2.6 Flash","maker":"Xiaomi","icon":"/agent-icons/mimo.svg","color":"var(--data-cyan)","pdr":64.3,"d0":null,"d1":26.1,"d2":57.3,"ttft":42.4,"time":958.1,"cacheHit":80.5,"inputTokens":1067362.5,"outputTokens":47656.2,"cost":0.02894,"calls":19.9,"scenarios":[{"id":"webapp","n":7,"score":93.9,"ci95":5,"pdr":100,"d1":28.9,"d2":65,"ttft":55.7,"time":1659,"cacheHit":85.2,"inputTokens":1732371.4,"outputTokens":76351.9,"cost":0.042654,"calls":34.2},{"id":"mobile","n":9,"score":75.4,"ci95":12.2,"pdr":44.4,"d1":23.3,"d2":52.1,"ttft":22.3,"time":808.3,"cacheHit":83.6,"inputTokens":1525112.1,"outputTokens":45483.6,"cost":0.028236,"calls":25.3},{"id":"desktop","n":8,"score":82.5,"ci95":5.8,"pdr":62.5,"d1":26.5,"d2":56,"ttft":30,"time":645.9,"cacheHit":74,"inputTokens":338405.7,"outputTokens":34536.8,"cost":0.01948,"calls":5.8},{"id":"dashboard","n":9,"score":84.4,"ci95":6,"pdr":66.7,"d1":26.1,"d2":58.3,"ttft":27,"time":586.3,"cacheHit":74.4,"inputTokens":494639.8,"outputTokens":32769.2,"cost":0.018441,"calls":7.4},{"id":"landing","n":9,"score":83.3,"ci95":7.6,"pdr":55.6,"d1":26.5,"d2":56.8,"ttft":76.8,"time":1091.3,"cacheHit":85.5,"inputTokens":1246283.7,"outputTokens":49139.8,"cost":0.035886,"calls":26.6}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:df380d7af56a8dd0","board":"open_design_arena","model":"gpt-6-luna","variant":"GPT-6 Luna","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":83.1,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.015871,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"gpt-6-luna","variantLabel":"GPT-6 Luna","score":83.1,"secondary":{"rank":5,"rankBasis":"average_score_desc","officialName":"GPT-6 Luna","maker":"OpenAI","icon":"/agents/openai.svg","color":"var(--data-cream)","pdr":70.5,"d0":null,"d1":25.2,"d2":57.9,"ttft":37.1,"time":254.9,"cacheHit":83.4,"inputTokens":316981.9,"outputTokens":17264.6,"cost":0.015871,"calls":9.2,"scenarios":[{"id":"webapp","n":9,"score":78.1,"ci95":7.1,"pdr":55.6,"d1":22.1,"d2":56,"ttft":29.3,"time":294.7,"cacheHit":84.2,"inputTokens":349631.7,"outputTokens":19267.6,"cost":0.017273,"calls":10.3},{"id":"mobile","n":9,"score":83.1,"ci95":3.8,"pdr":66.7,"d1":27.1,"d2":56,"ttft":46.4,"time":237.3,"cacheHit":85.7,"inputTokens":373028.3,"outputTokens":18067.1,"cost":0.016973,"calls":10.4},{"id":"desktop","n":8,"score":80.5,"ci95":10.2,"pdr":62.5,"d1":23.7,"d2":56.9,"ttft":34.8,"time":237.7,"cacheHit":81.7,"inputTokens":227585.6,"outputTokens":16816.7,"cost":0.014342,"calls":6.7},{"id":"dashboard","n":9,"score":87.2,"ci95":7.3,"pdr":77.8,"d1":25.7,"d2":61.4,"ttft":32.2,"time":231.7,"cacheHit":82.8,"inputTokens":245478.6,"outputTokens":16381.9,"cost":0.014116,"calls":7},{"id":"landing","n":9,"score":86.5,"ci95":5,"pdr":88.9,"d1":27.3,"d2":59.1,"ttft":42.6,"time":273,"cacheHit":82.3,"inputTokens":389185.2,"outputTokens":15790,"cost":0.016652,"calls":11.7}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:91957ddea79bc27b","board":"open_design_arena","model":"grok-4.7","variant":"Grok 4.7","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":82.9,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.662291,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"grok-4.7","variantLabel":"Grok 4.7","score":82.9,"secondary":{"rank":6,"rankBasis":"average_score_desc","officialName":"Grok 4.7","maker":"xAI","icon":"/model-icons/grok.svg","color":"var(--data-coral)","pdr":73.8,"d0":null,"d1":25.9,"d2":57,"ttft":50.6,"time":977.2,"cacheHit":68.6,"inputTokens":458694.9,"outputTokens":44267.5,"cost":0.662291,"calls":12.9,"scenarios":[{"id":"webapp","n":8,"score":88.1,"ci95":4.3,"pdr":87.5,"d1":26,"d2":62.1,"ttft":45.2,"time":969.6,"cacheHit":71.3,"inputTokens":479269.9,"outputTokens":37448.1,"cost":0.647548,"calls":13.4},{"id":"mobile","n":9,"score":85.8,"ci95":7.4,"pdr":55.6,"d1":29.1,"d2":56.8,"ttft":42,"time":831,"cacheHit":66.1,"inputTokens":399704.6,"outputTokens":39499.1,"cost":0.601801,"calls":12.8},{"id":"desktop","n":8,"score":79.7,"ci95":16.1,"pdr":75,"d1":22.8,"d2":56.9,"ttft":47.2,"time":995.3,"cacheHit":68.3,"inputTokens":309055.3,"outputTokens":54658.2,"cost":0.621047,"calls":9.7},{"id":"dashboard","n":9,"score":79.1,"ci95":16.6,"pdr":77.8,"d1":23.1,"d2":56,"ttft":45,"time":1177.2,"cacheHit":68.8,"inputTokens":293074.9,"outputTokens":48102.7,"cost":0.550606,"calls":7.6},{"id":"landing","n":8,"score":81.6,"ci95":3.9,"pdr":75,"d1":28.3,"d2":53.4,"ttft":73.8,"time":912.8,"cacheHit":68.5,"inputTokens":812370,"outputTokens":41629.4,"cost":0.890453,"calls":21}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:eab6a68a5b32527b","board":"open_design_arena","model":"gpt-6-astra","variant":"GPT-6 Astra","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":82.7,"score_low":null,"score_high":null,"mean_cost_usd_per_task":1.612939,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"gpt-6-astra","variantLabel":"GPT-6 Astra","score":82.7,"secondary":{"rank":7,"rankBasis":"average_score_desc","officialName":"GPT-6 Astra","maker":"OpenAI","icon":"/agents/openai.svg","color":"var(--data-teal)","pdr":60,"d0":null,"d1":26.5,"d2":56.2,"ttft":13,"time":663,"cacheHit":81,"inputTokens":318676,"outputTokens":15544,"cost":1.612939,"calls":4.3,"scenarios":[{"id":"webapp","n":6,"score":77.1,"ci95":7.8,"pdr":33.3,"d1":26.9,"d2":50.2,"ttft":13,"time":907,"cacheHit":84,"inputTokens":357347,"outputTokens":20301,"calls":5.3,"cost":1.871887},{"id":"mobile","n":6,"score":87.4,"ci95":12.3,"pdr":66.7,"d1":25.6,"d2":61.8,"ttft":13,"time":555,"cacheHit":82.6,"inputTokens":313900,"outputTokens":11873,"calls":4.5,"cost":1.39985},{"id":"desktop","n":6,"score":86.3,"ci95":4.2,"pdr":83.3,"d1":26.8,"d2":59.5,"ttft":12,"time":586,"cacheHit":81.4,"inputTokens":271110,"outputTokens":15814,"calls":3.7,"cost":1.514219},{"id":"dashboard","n":6,"score":86.3,"ci95":12.2,"pdr":66.7,"d1":27.9,"d2":58.3,"ttft":13,"time":748,"cacheHit":74.1,"inputTokens":244815,"outputTokens":16917,"calls":3.3,"cost":1.652953},{"id":"landing","n":6,"score":76.6,"ci95":13.1,"pdr":50,"d1":25.2,"d2":51.3,"ttft":14,"time":518,"cacheHit":82.9,"inputTokens":406206,"outputTokens":12818,"calls":4.7,"cost":1.625788}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:050e7234636e98be","board":"open_design_arena","model":"deepseek-v4.1-flash","variant":"DeepSeek V4.1 Flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":81.2,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.023264,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"deepseek-v4.1-flash","variantLabel":"DeepSeek V4.1 Flash","score":81.2,"secondary":{"rank":8,"rankBasis":"average_score_desc","officialName":"DeepSeek V4.1 Flash","maker":"DeepSeek","icon":"/agent-icons/deepseek.svg","color":"var(--data-cyan)","pdr":57.7,"d0":null,"d1":28.4,"d2":52.8,"ttft":38,"time":316,"cacheHit":89,"inputTokens":1462767,"outputTokens":29202,"cost":0.023264,"calls":18.9,"scenarios":[{"id":"webapp","n":6,"score":82.4,"ci95":4.7,"pdr":66.7,"d1":28.8,"d2":53.7,"ttft":33,"time":540,"cacheHit":90.3,"inputTokens":2549279,"outputTokens":40102,"calls":27,"cost":0.029765},{"id":"mobile","n":4,"score":82.5,"ci95":20.6,"pdr":50,"d1":30,"d2":52.5,"ttft":60,"time":304,"cacheHit":91.4,"inputTokens":1789582,"outputTokens":27341,"calls":23.8,"cost":0.025778},{"id":"desktop","n":6,"score":84.4,"ci95":7.4,"pdr":66.7,"d1":28.4,"d2":56,"ttft":34,"time":192,"cacheHit":82.6,"inputTokens":525575,"outputTokens":25412,"calls":7.5,"cost":0.018847},{"id":"dashboard","n":4,"score":76.1,"ci95":16.8,"pdr":75,"d1":27.1,"d2":49,"ttft":20,"time":151,"cacheHit":87,"inputTokens":559389,"outputTokens":26250,"calls":8,"cost":0.016186},{"id":"landing","n":6,"score":79.3,"ci95":7.3,"pdr":33.3,"d1":27.9,"d2":51.3,"ttft":43,"time":334,"cacheHit":93.9,"inputTokens":1697823,"outputTokens":25300,"calls":26.2,"cost":0.024223}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:0ed675e0b61ae3ac","board":"open_design_arena","model":"claude-fable-5.1","variant":"Claude Fable 5.1","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":80.3,"score_low":null,"score_high":null,"mean_cost_usd_per_task":3.660199,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"claude-fable-5.1","variantLabel":"Claude Fable 5.1","score":80.3,"secondary":{"rank":9,"rankBasis":"average_score_desc","officialName":"Claude Fable 5.1","maker":"Anthropic","icon":"/model-icons/claude-brand.svg","color":"var(--data-cream)","pdr":56.7,"d0":null,"d1":27.1,"d2":53.2,"ttft":41,"time":766,"cacheHit":94,"inputTokens":715653,"outputTokens":63610,"cost":3.660199,"calls":9,"scenarios":[{"id":"webapp","n":6,"score":82.6,"ci95":3.1,"pdr":66.7,"d1":27.8,"d2":54.8,"ttft":38,"time":1144,"cacheHit":95,"inputTokens":991255,"outputTokens":100273,"calls":11.5,"cost":5.552063},{"id":"mobile","n":6,"score":79,"ci95":20.7,"pdr":66.7,"d1":30,"d2":49,"ttft":46,"time":843,"cacheHit":94.3,"inputTokens":686093,"outputTokens":68028,"calls":8.8,"cost":3.876613},{"id":"desktop","n":6,"score":73.3,"ci95":11.3,"pdr":33.3,"d1":20.8,"d2":52.5,"ttft":38,"time":724,"cacheHit":93.1,"inputTokens":462756,"outputTokens":66216,"calls":5.8,"cost":3.716984},{"id":"dashboard","n":6,"score":85.4,"ci95":6.2,"pdr":66.7,"d1":28.2,"d2":57.2,"ttft":39,"time":566,"cacheHit":92.5,"inputTokens":462564,"outputTokens":49259,"calls":5.7,"cost":2.869084},{"id":"landing","n":6,"score":81.3,"ci95":9.9,"pdr":50,"d1":28.8,"d2":52.5,"ttft":45,"time":551,"cacheHit":94.9,"inputTokens":975598,"outputTokens":34275,"calls":13.2,"cost":2.286252}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:e7590d5e99a8809f","board":"open_design_arena","model":"glm-5.3-flashx","variant":"GLM-5.3 Flash-X","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":80.1,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.097833,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"glm-5.3-flashx","variantLabel":"GLM-5.3 Flash-X","score":80.1,"secondary":{"rank":10,"rankBasis":"average_score_desc","officialName":"GLM-5.3 Flash-X","maker":"Z.ai","icon":"/model-icons/zhipu.svg","color":"var(--data-coral)","pdr":53.7,"d0":null,"d1":27,"d2":53.1,"ttft":40,"time":478.6,"cacheHit":66.6,"inputTokens":390547.4,"outputTokens":32934.6,"cost":0.097833,"calls":6.9,"scenarios":[{"id":"webapp","n":8,"score":78,"ci95":7.8,"pdr":62.5,"d1":26.4,"d2":51.6,"ttft":37.1,"time":379.5,"cacheHit":66.6,"inputTokens":307514,"outputTokens":37735.8,"cost":0.097186,"calls":5.2},{"id":"mobile","n":8,"score":74.6,"ci95":7.8,"pdr":25,"d1":30,"d2":44.6,"ttft":42,"time":305,"cacheHit":65,"inputTokens":377286.2,"outputTokens":37414.2,"cost":0.105169,"calls":7.6},{"id":"desktop","n":8,"score":81.3,"ci95":5.8,"pdr":62.5,"d1":25.3,"d2":56,"ttft":36.6,"time":222.4,"cacheHit":67.1,"inputTokens":324610.4,"outputTokens":28958.4,"cost":0.087086,"calls":4.9},{"id":"dashboard","n":9,"score":82.5,"ci95":4.5,"pdr":66.7,"d1":24.9,"d2":57.6,"ttft":37.5,"time":238.2,"cacheHit":52.4,"inputTokens":215722.1,"outputTokens":29261.7,"cost":0.07756,"calls":2.7},{"id":"landing","n":8,"score":83.7,"ci95":6.9,"pdr":50,"d1":28.5,"d2":55.1,"ttft":47,"time":1248.1,"cacheHit":81.8,"inputTokens":727604.1,"outputTokens":31302.9,"cost":0.122162,"calls":14.1}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:048a6b75c77d30fe","board":"open_design_arena","model":"gpt-5.6-sol","variant":"GPT-5.6 Sol","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":77.6,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.544242,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"gpt-5.6-sol","variantLabel":"GPT-5.6 Sol","score":77.6,"secondary":{"rank":11,"rankBasis":"average_score_desc","officialName":"GPT-5.6 Sol","maker":"OpenAI","icon":"/agents/openai.svg","color":"var(--data-teal)","pdr":53.3,"d0":100,"d1":25.8,"d2":51.8,"ttft":19,"time":193,"cacheHit":80.7,"inputTokens":236582,"outputTokens":14521,"cost":0.544242,"calls":5.9,"scenarios":[{"id":"webapp","n":6,"score":83.1,"ci95":10.2,"pdr":66.7,"d1":25.9,"d2":57.2,"ttft":19,"time":203,"cacheHit":81,"inputTokens":232352,"outputTokens":14501,"cost":0.53749,"calls":6},{"id":"mobile","n":6,"score":86.6,"ci95":7.6,"pdr":66.7,"d1":29.4,"d2":57.2,"ttft":17,"time":231,"cacheHit":80.7,"inputTokens":246562,"outputTokens":15483,"cost":0.577312,"calls":6.3},{"id":"desktop","n":6,"score":79.8,"ci95":10.2,"pdr":50,"d1":23.8,"d2":56,"ttft":16,"time":169,"cacheHit":80.1,"inputTokens":213814,"outputTokens":14450,"cost":0.526577,"calls":5.3},{"id":"dashboard","n":6,"score":79,"ci95":13.2,"pdr":50,"d1":24.2,"d2":54.8,"ttft":20,"time":190,"cacheHit":78.3,"inputTokens":200709,"outputTokens":15454,"cost":0.545071,"calls":4.8},{"id":"landing","n":6,"score":59.4,"ci95":24.2,"pdr":33.3,"d1":25.5,"d2":33.8,"ttft":24,"time":171,"cacheHit":83.4,"inputTokens":289472,"outputTokens":12715,"cost":0.534759,"calls":7.2}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:1d059a6d807d4308","board":"open_design_arena","model":"hy4-preview","variant":"Hunyuan H4 Preview","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":74.6,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.417793,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"hy4-preview","variantLabel":"Hunyuan H4 Preview","score":74.6,"secondary":{"rank":12,"rankBasis":"average_score_desc","officialName":"Hunyuan H4 Preview","maker":"Tencent","icon":"/model-icons/hunyuan.svg","color":"var(--data-cyan)","pdr":40,"d0":null,"d1":26.8,"d2":47.8,"ttft":129,"time":1752,"cacheHit":90.9,"inputTokens":3085880,"outputTokens":45980,"cost":0.417793,"calls":34.9,"scenarios":[{"id":"webapp","n":6,"score":60.5,"ci95":24.3,"pdr":16.7,"d1":22,"d2":38.5,"ttft":227,"time":2423,"cacheHit":92.4,"inputTokens":4999340,"outputTokens":63800,"calls":54,"cost":0.609047},{"id":"mobile","n":6,"score":89.5,"ci95":10.5,"pdr":83.3,"d1":30,"d2":59.5,"ttft":109,"time":1689,"cacheHit":91.3,"inputTokens":2889220,"outputTokens":42785,"calls":33.5,"cost":0.404329},{"id":"desktop","n":6,"score":74.7,"ci95":8.2,"pdr":33.3,"d1":28,"d2":46.7,"ttft":74,"time":1506,"cacheHit":90.6,"inputTokens":2688808,"outputTokens":44722,"calls":28,"cost":0.369551},{"id":"dashboard","n":6,"score":80.3,"ci95":12.5,"pdr":50,"d1":26.7,"d2":53.7,"ttft":105,"time":1458,"cacheHit":88.1,"inputTokens":2446914,"outputTokens":44259,"calls":25.5,"cost":0.372523},{"id":"landing","n":6,"score":67.9,"ci95":22,"pdr":16.7,"d1":27.1,"d2":40.8,"ttft":132,"time":1684,"cacheHit":92,"inputTokens":2405116,"outputTokens":34335,"calls":33.5,"cost":0.333516}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:fbd3d33fa1d7cf9a","board":"open_design_arena","model":"qwen3.8-max","variant":"Qwen 3.8-Max","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":"max","service_mode":null,"score":72,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.595399,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"qwen3.8-max","variantLabel":"Qwen 3.8-Max","score":72,"secondary":{"rank":13,"rankBasis":"average_score_desc","officialName":"Qwen 3.8-Max","maker":"Alibaba","icon":"/agent-icons/qwen.svg","color":"var(--data-cream)","pdr":56.7,"d0":90,"d1":23,"d2":49,"ttft":227,"time":1587,"cacheHit":79,"inputTokens":836797,"outputTokens":28629,"cost":0.595399,"calls":12,"scenarios":[{"id":"webapp","n":6,"score":57,"ci95":33.2,"pdr":50,"d1":18.5,"d2":38.5,"ttft":223,"time":1391,"cacheHit":75,"inputTokens":763100,"outputTokens":27770,"cost":0.552174,"calls":11.3},{"id":"mobile","n":6,"score":56.7,"ci95":27.8,"pdr":33.3,"d1":21.7,"d2":35,"ttft":270,"time":2103,"cacheHit":84,"inputTokens":1551556,"outputTokens":39009,"cost":0.919458,"calls":20.2},{"id":"desktop","n":6,"score":81.4,"ci95":10.4,"pdr":66.7,"d1":24.3,"d2":57.2,"ttft":197,"time":1325,"cacheHit":76.5,"inputTokens":441560,"outputTokens":27448,"cost":0.451447,"calls":7.3},{"id":"dashboard","n":6,"score":82.4,"ci95":7.6,"pdr":66.7,"d1":24,"d2":58.3,"ttft":192,"time":1285,"cacheHit":77.1,"inputTokens":474481,"outputTokens":24405,"cost":0.445005,"calls":6.8},{"id":"landing","n":6,"score":82.8,"ci95":6.3,"pdr":66.7,"d1":26.8,"d2":56,"ttft":254,"time":1832,"cacheHit":82.3,"inputTokens":953289,"outputTokens":24516,"cost":0.60891,"calls":14.5}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:e5b1ca76e59dabd0","board":"open_design_arena","model":"gemini-3.8-flash","variant":"Gemini 3.8 Flash","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":68.9,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.210201,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"gemini-3.8-flash","variantLabel":"Gemini 3.8 Flash","score":68.9,"secondary":{"rank":14,"rankBasis":"average_score_desc","officialName":"Gemini 3.8 Flash","maker":"Google","icon":"/model-icons/gemini-brand.svg","color":"var(--data-coral)","pdr":30,"d0":null,"d1":25,"d2":43.9,"ttft":34,"time":227,"cacheHit":70.9,"inputTokens":392340,"outputTokens":27851,"cost":0.210201,"calls":7.1,"scenarios":[{"id":"webapp","n":6,"score":68.4,"ci95":28.6,"pdr":33.3,"d1":21.7,"d2":46.7,"ttft":24,"time":179,"cacheHit":74.4,"inputTokens":365916,"outputTokens":25620,"calls":7.3,"cost":0.185069},{"id":"mobile","n":6,"score":79.4,"ci95":17.5,"pdr":50,"d1":28.1,"d2":51.3,"ttft":36,"time":217,"cacheHit":74.3,"inputTokens":463384,"outputTokens":27386,"calls":8.2,"cost":0.217424},{"id":"desktop","n":6,"score":75.6,"ci95":8.1,"pdr":16.7,"d1":26.6,"d2":49,"ttft":42,"time":223,"cacheHit":66.8,"inputTokens":371074,"outputTokens":29013,"calls":6.8,"cost":0.21935},{"id":"dashboard","n":6,"score":74.7,"ci95":15.9,"pdr":33.3,"d1":29.2,"d2":45.5,"ttft":23,"time":259,"cacheHit":69.5,"inputTokens":319569,"outputTokens":26921,"calls":6,"cost":0.190314},{"id":"landing","n":6,"score":46.5,"ci95":27.8,"pdr":16.7,"d1":19.7,"d2":26.8,"ttft":45,"time":257,"cacheHit":69.4,"inputTokens":441758,"outputTokens":30316,"calls":7.3,"cost":0.238847}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:3d2b13f88b32b8ff","board":"open_design_arena","model":"muse-spark-1.3","variant":"Muse Spark 1.3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":66.6,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.267309,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"muse-spark-1.3","variantLabel":"Muse Spark 1.3","score":66.6,"secondary":{"rank":15,"rankBasis":"average_score_desc","officialName":"Muse Spark 1.3","maker":"Meta","icon":"/model-icons/meta.png","color":"var(--data-teal)","pdr":35.7,"d0":null,"d1":25.6,"d2":41,"ttft":30,"time":198,"cacheHit":31.5,"inputTokens":235926,"outputTokens":13428,"cost":0.267309,"calls":5.1,"scenarios":[{"id":"webapp","n":6,"score":62,"ci95":8.1,"pdr":0,"d1":25.8,"d2":36.2,"ttft":28,"time":242,"cacheHit":28.2,"inputTokens":245696,"outputTokens":14896,"calls":5.3,"cost":0.29313},{"id":"mobile","n":4,"score":63.3,"ci95":15.4,"pdr":25,"d1":21.3,"d2":42,"ttft":40,"time":191,"cacheHit":32.9,"inputTokens":253534,"outputTokens":12867,"calls":6,"cost":0.273529},{"id":"desktop","n":6,"score":76.4,"ci95":16.2,"pdr":66.7,"d1":27.4,"d2":49,"ttft":26,"time":176,"cacheHit":22.7,"inputTokens":206906,"outputTokens":13760,"calls":4,"cost":0.266792},{"id":"dashboard","n":6,"score":76.8,"ci95":17.5,"pdr":66.7,"d1":26.7,"d2":50.2,"ttft":29,"time":177,"cacheHit":32.9,"inputTokens":230331,"outputTokens":13922,"calls":5,"cost":0.263141},{"id":"landing","n":6,"score":53.4,"ci95":23,"pdr":16.7,"d1":25.4,"d2":28,"ttft":31,"time":199,"cacheHit":41.2,"inputTokens":249035,"outputTokens":11507,"calls":5.3,"cost":0.242027}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}},{"configuration_id":"open_design_arena:9aadc4d724514f03","board":"open_design_arena","model":"kimi-k3","variant":"Kimi K3","score_is_estimated":null,"score_is_self_reported":false,"agent_harness":"OpenDesign","reasoning_effort":null,"service_mode":null,"score":65.7,"score_low":null,"score_high":null,"mean_cost_usd_per_task":0.613653,"median_cost_usd_per_task":null,"source":"https://open-design.ai/zh/llm-arena-for-design/","checked_at":"2026-10-01","archive":"scores-open-design-round2-2026-10-01.json","raw_record":{"boardId":"open_design_arena","model":"kimi-k3","variantLabel":"Kimi K3","score":65.7,"secondary":{"rank":16,"rankBasis":"average_score_desc","officialName":"Kimi K3","maker":"Moonshot AI","icon":"/model-icons/arena-kimi.svg","color":"var(--data-cyan)","pdr":38.7,"d0":90.3,"d1":22.4,"d2":43.4,"ttft":137,"time":841,"cacheHit":71.7,"inputTokens":427094,"outputTokens":16066,"cost":0.613653,"calls":9.1,"scenarios":[{"id":"webapp","n":7,"score":35.9,"ci95":23.8,"pdr":0,"d1":12.9,"d2":23,"ttft":114,"time":595,"cacheHit":57.5,"inputTokens":355722,"outputTokens":12252,"cost":0.485737,"calls":8.2},{"id":"mobile","n":6,"score":80.6,"ci95":11.8,"pdr":66.7,"d1":26.9,"d2":53.7,"ttft":192,"time":1242,"cacheHit":72.9,"inputTokens":728579,"outputTokens":21926,"cost":1.051856,"calls":15.5},{"id":"desktop","n":6,"score":84.4,"ci95":3.7,"pdr":83.3,"d1":24.9,"d2":59.5,"ttft":99,"time":742,"cacheHit":68.9,"inputTokens":260284,"outputTokens":17984,"cost":0.537769,"calls":5},{"id":"dashboard","n":6,"score":68.1,"ci95":10.5,"pdr":16.7,"d1":27.2,"d2":40.8,"ttft":147,"time":627,"cacheHit":76.7,"inputTokens":244405,"outputTokens":13813,"cost":0.434424,"calls":4.3},{"id":"landing","n":6,"score":64.5,"ci95":20.7,"pdr":33.3,"d1":21.4,"d2":43.2,"ttft":137,"time":1041,"cacheHit":84.9,"inputTokens":558374,"outputTokens":14991,"cost":0.579798,"calls":12.7}]},"source":"https://open-design.ai/zh/llm-arena-for-design/","checkedAt":"2026-10-01"}}]