{"benchmarks":[{"id":"gpqa-diamond-aa","title":"Best in Reasoning","label":"GPQA Diamond","scope":"198-question Diamond subset; independent AA evaluation","note":"Independent GPQA Diamond results. Best reported effort per model; full precision retained.","publisher":"Artificial Analysis","url":"https://artificialanalysis.ai/evaluations/gpqa-diamond","precision":1,"unit":"%","scaleMin":0,"scaleMax":100,"slotId":"gpqa-diamond-aa","models":[{"benchmark_id":"gpqa-diamond-aa","id":"kimi-k3","name":"Kimi K3","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":93.53535353535351,"source":"https://artificialanalysis.ai/models/kimi-k3","settings":"Kimi K3 (Max); 198-question Diamond subset; independent AA evaluation","model_rank":1,"benchmark_rank":1},{"benchmark_id":"gpqa-diamond-aa","id":"minimax-m3","name":"MiniMax M3","provider":"MiniMax","providerId":"minimax","logo":"/providers/minimax.ico","score":92.9292929292929,"source":"https://artificialanalysis.ai/models/minimax-m3","settings":"MiniMax-M3; 198-question Diamond subset; independent AA evaluation","model_rank":1,"benchmark_rank":2},{"benchmark_id":"gpqa-diamond-aa","id":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":92.82828282828281,"source":"https://artificialanalysis.ai/models/deepseek-v4-pro","settings":"DeepSeek V4 Pro 0813 (Max); 198-question Diamond subset; independent AA evaluation","model_rank":1,"benchmark_rank":3},{"benchmark_id":"gpqa-diamond-aa","id":"glm-5-3","name":"GLM-5.3","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":91.7171717171717,"source":"https://artificialanalysis.ai/models/glm-5-3","settings":"GLM-5.3 (Max); 198-question Diamond subset; independent AA evaluation","model_rank":1,"benchmark_rank":4},{"benchmark_id":"gpqa-diamond-aa","id":"glm-5-3-flash","name":"GLM-5.3-Flash","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":91.2121212121212,"source":"https://artificialanalysis.ai/models/glm-5-3-flash","settings":"GLM-5.3-Flash; 198-question Diamond subset; independent AA evaluation","model_rank":1,"benchmark_rank":5}]},{"id":"swe-verified-open-reports","title":"Best in Coding","label":"SWE-bench Verified","scope":"Verified split; published reports","note":"Provider-reported scores; agent harnesses differ. Every row retains its harness and original source.","publisher":"Model publishers","url":"https://www.swebench.com/","precision":1,"unit":"%","scaleMin":0,"scaleMax":100,"slotId":"swe-verified-open-reports","models":[{"benchmark_id":"swe-verified-open-reports","id":"deepseek-v4-pro-0424","name":"DeepSeek V4 Pro 0424","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":80.6,"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro","settings":"DeepSeek original V4 release; Think Max; published SWE Verified resolved score; original release, not 0813","model_rank":1,"benchmark_rank":1},{"benchmark_id":"swe-verified-open-reports","id":"minimax-m3","name":"MiniMax M3","provider":"MiniMax","providerId":"minimax","logo":"/providers/minimax.ico","score":80.5,"source":"https://www.minimax.io/blog/minimax-m3","settings":"MiniMax primary benchmark table; internal Claude Code harness; system prompt override; average of four runs","model_rank":1,"benchmark_rank":2},{"benchmark_id":"swe-verified-open-reports","id":"kimi-k2-6","name":"Kimi K2.6","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":80.2,"source":"https://www.minimax.io/blog/minimax-m3","settings":"Kimi K2.6 Thinking; quoted comparison table; follow original Kimi SWE harness","model_rank":1,"benchmark_rank":3},{"benchmark_id":"swe-verified-open-reports","id":"deepseek-v4-flash-0420","name":"DeepSeek V4 Flash 0420","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":79,"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro","settings":"DeepSeek original V4 release; Think Max; published SWE Verified resolved score; original release, not 0813","model_rank":1,"benchmark_rank":4},{"benchmark_id":"swe-verified-open-reports","id":"kimi-k2-5","name":"Kimi K2.5","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":76.8,"source":"https://huggingface.co/moonshotai/Kimi-K2.5","settings":"Internal minimal coding agent; non-thinking; average of five independent runs","model_rank":1,"benchmark_rank":5}]},{"id":"terminal-bench-2-1-aa","title":"Best in Terminal Use","label":"Terminal-Bench 2.1","scope":"89 tasks; Terminus 2; e2b; three repeats","note":"Independent AA evaluation of Terminal-Bench 2.1. Never mix 2.0 or 4.0 scores.","publisher":"Artificial Analysis","url":"https://artificialanalysis.ai/evaluations/terminalbench-2-1","precision":1,"unit":"%","scaleMin":0,"scaleMax":100,"slotId":"terminal-bench-2-1-aa","models":[{"benchmark_id":"terminal-bench-2-1-aa","id":"kimi-k3","name":"Kimi K3","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":85.0187265917603,"source":"https://artificialanalysis.ai/models/kimi-k3","settings":"Kimi K3 (Max); 89 tasks; Terminus 2; e2b; three repeats","model_rank":1,"benchmark_rank":1},{"benchmark_id":"terminal-bench-2-1-aa","id":"glm-5-3-flash","name":"GLM-5.3-Flash","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":84.2696629213483,"source":"https://artificialanalysis.ai/models/glm-5-3-flash","settings":"GLM-5.3-Flash; 89 tasks; Terminus 2; e2b; three repeats","model_rank":1,"benchmark_rank":2},{"benchmark_id":"terminal-bench-2-1-aa","id":"glm-5-3","name":"GLM-5.3","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":83.8951310861423,"source":"https://artificialanalysis.ai/models/glm-5-3","settings":"GLM-5.3 (Max); 89 tasks; Terminus 2; e2b; three repeats","model_rank":1,"benchmark_rank":3},{"benchmark_id":"terminal-bench-2-1-aa","id":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":78.65168539325839,"source":"https://artificialanalysis.ai/models/deepseek-v4-pro","settings":"DeepSeek V4 Pro 0813 (Max); 89 tasks; Terminus 2; e2b; three repeats","model_rank":1,"benchmark_rank":4},{"benchmark_id":"terminal-bench-2-1-aa","id":"glm-5-2","name":"GLM-5.2","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":77.9026217228464,"source":"https://artificialanalysis.ai/models/glm-5-2","settings":"GLM-5.2 (Max); 89 tasks; Terminus 2; e2b; three repeats","model_rank":1,"benchmark_rank":5}]},{"id":"browsecomp-open-reports","title":"Best in Web Research","label":"BrowseComp","scope":"Published reports; source settings retained","note":"Provider-reported scores; browsing tools, budgets and context-management settings differ. This ranking is not a matched-harness comparison.","publisher":"Model publishers","url":"https://openai.com/index/browsecomp/","precision":1,"unit":"%","scaleMin":0,"scaleMax":100,"slotId":"browsecomp-open-reports","models":[{"benchmark_id":"browsecomp-open-reports","id":"kimi-k3-1-pro","name":"Kimi K3.1 Pro","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":92.1,"source":"https://www.kimi.ai/blog/kimi-k3-1-pro","settings":"200K context compaction; max 500 steps; temperature 1; top-p 1; preserved original track kimi31-report-200k-500steps","model_rank":1,"benchmark_rank":1},{"benchmark_id":"browsecomp-open-reports","id":"kimi-k3","name":"Kimi K3","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":91.2,"source":"https://www.kimi.ai/blog/kimi-k3","settings":"Max effort; temperature 1; top_p 1; context compaction triggered at 300K tokens. Primary General Agents benchmark image. Separate no-management 1M track scores 90.4.","model_rank":1,"benchmark_rank":2},{"benchmark_id":"browsecomp-open-reports","id":"minimax-m3","name":"MiniMax M3","provider":"MiniMax","providerId":"minimax","logo":"/providers/minimax.ico","score":83.52,"source":"https://www.minimax.io/blog/minimax-m3","settings":"Primary full table; WebExplorer-based agent; discard history after 64K tokens","model_rank":1,"benchmark_rank":3},{"benchmark_id":"browsecomp-open-reports","id":"deepseek-v4-pro-0424","name":"DeepSeek V4 Pro 0424","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":83.4,"source":"https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro","settings":"Think Max; provider browsing agent; discard-all context management; original V4 release","model_rank":1,"benchmark_rank":4},{"benchmark_id":"browsecomp-open-reports","id":"kimi-k2-6","name":"Kimi K2.6","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":83.2,"source":"https://forum.moonshot.ai/t/meet-kimi-k2-6-advancing-open-source-coding/369","settings":"Moonshot provider-reported tool-augmented BrowseComp","model_rank":1,"benchmark_rank":5}]},{"id":"automationbench-aa-1-0-6","title":"Best in Work Automations","label":"AutomationBench-AA","scope":"Private 657 tasks; 50 turns; objective completion","note":"Independent AA objective-completion scores, not Zapier strict task pass/fail scores.","publisher":"Artificial Analysis","url":"https://artificialanalysis.ai/evaluations/automationbench","precision":2,"unit":"%","scaleMin":0,"scaleMax":100,"slotId":"automationbench-aa-1-0-6","models":[{"benchmark_id":"automationbench-aa-1-0-6","id":"kimi-k3-1-pro","name":"Kimi K3.1 Pro","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":73.3,"source":"https://artificialanalysis.ai/evaluations/automationbench-aa","settings":"Artificial Analysis evaluation setup","model_rank":1,"benchmark_rank":1},{"benchmark_id":"automationbench-aa-1-0-6","id":"minimax-m3-1-flash","name":"MiniMax M3.1 Flash","provider":"MiniMax","providerId":"minimax","logo":"/providers/minimax.ico","score":72.52180957532543,"source":"https://artificialanalysis.ai/models/minimax-m3-1-flash","settings":"MiniMax-M3.1-Flash; Private 657 tasks; 50 turns; objective completion","model_rank":1,"benchmark_rank":2},{"benchmark_id":"automationbench-aa-1-0-6","id":"deepseek-v4-1-flash","name":"DeepSeek V4.1 Flash","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":68.89097769674058,"source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash","settings":"DeepSeek V4.1 Flash (Max); Private 657 tasks; 50 turns; objective completion","model_rank":1,"benchmark_rank":3},{"benchmark_id":"automationbench-aa-1-0-6","id":"deepseek-v4-1-flash-1017","name":"DeepSeek V4.1 Flash 1017","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":68.49296525816153,"source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash-1017-high","settings":"DeepSeek V4.1 Flash 1017 (High); Private 657 tasks; 50 turns; objective completion","model_rank":1,"benchmark_rank":4},{"benchmark_id":"automationbench-aa-1-0-6","id":"glm-5-3","name":"GLM-5.3","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":62.202864996264196,"source":"https://artificialanalysis.ai/models/glm-5-3","settings":"GLM-5.3 (Max); Private 657 tasks; 50 turns; objective completion","model_rank":1,"benchmark_rank":5}]},{"id":"gdpval-aa-v2-1","title":"Best in Knowledge Work","label":"GDPval-AA v2.1","scope":"220 tasks; Stirrup agent; pairwise Elo","note":"Independent v2.1 Elo ratings, not percentages or v2.0 ratings. Best reported effort per model.","publisher":"Artificial Analysis","url":"https://artificialanalysis.ai/evaluations/gdpval-aa","precision":0,"unit":"Elo","scaleMin":0,"scaleMax":2000,"slotId":"gdpval-aa-v2-1","models":[{"benchmark_id":"gdpval-aa-v2-1","id":"kimi-k3-1-pro","name":"Kimi K3.1 Pro","provider":"Moonshot AI","providerId":"moonshot-ai","logo":"/providers/moonshot-ai.svg","score":1858.05,"source":"https://artificialanalysis.ai/models/kimi-k3-1-pro","settings":"Kimi K3.1 Pro; 220 tasks; Stirrup agent; pairwise Elo","model_rank":1,"benchmark_rank":1},{"benchmark_id":"gdpval-aa-v2-1","id":"deepseek-v4-1-flash-1017","name":"DeepSeek V4.1 Flash 1017","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":1745.77,"source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash-1017","settings":"DeepSeek V4.1 Flash 1017 (Max); 220 tasks; Stirrup agent; pairwise Elo","model_rank":1,"benchmark_rank":2},{"benchmark_id":"gdpval-aa-v2-1","id":"glm-5-3","name":"GLM-5.3","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":1656.88,"source":"https://artificialanalysis.ai/models/glm-5-3","settings":"GLM-5.3 (Max); 220 tasks; Stirrup agent; pairwise Elo","model_rank":1,"benchmark_rank":3},{"benchmark_id":"gdpval-aa-v2-1","id":"glm-5-3-flash","name":"GLM-5.3-Flash","provider":"Z.ai","providerId":"zai","logo":"/providers/zai.svg","score":1649.37,"source":"https://artificialanalysis.ai/models/glm-5-3-flash","settings":"GLM-5.3-Flash; 220 tasks; Stirrup agent; pairwise Elo","model_rank":1,"benchmark_rank":4},{"benchmark_id":"gdpval-aa-v2-1","id":"deepseek-v4-1-flash","name":"DeepSeek V4.1 Flash","provider":"DeepSeek","providerId":"deepseek","logo":"/providers/deepseek.svg","score":1600,"source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash","settings":"DeepSeek V4.1 Flash (Max); 220 tasks; Stirrup agent; pairwise Elo","model_rank":1,"benchmark_rank":5}]}]}