BenchmarkActivecategory · agentic
quality57
Updated 3 h ago · first seen 11 Sept 2026
bench_01M293SPFPKK4MF90JEDK0PH8C
- Metric
- accuracy · %
- Direction
- Higher is better
- Results
- 821 · 432 filtered
- Leader
- Claude Fable 5.1 91.39%
Leaderboard 432 current results · config contains “hard”
Select models with +, then open Compare.
| # | Model | Score | Config | Evaluated | Source | Actions |
|---|---|---|---|---|---|---|
| #1gpt-5-6-solOpenAI | 65.91% | aa_slug=gpt-5-6-sol · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #2gpt-5-6-sol-mediumOpenAI | 62.88% | aa_slug=gpt-5-6-sol-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #3gpt-5-6-terra-xhighOpenAI | 62.88% | aa_slug=gpt-5-6-terra-xhigh · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #4Claude Fable 5Anthropic | 62.88% | aa_slug=claude-fable-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #5gpt-5-6-sol-highOpenAI | 62.12% | aa_slug=gpt-5-6-sol-high · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #6gpt-5-6-sol-xhighOpenAI | 61.36% | aa_slug=gpt-5-6-sol-xhigh · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #7gpt-5.5OpenAI | 60.61% | aa_slug=gpt-5-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #8gpt-5-6-sol-lowOpenAI | 60.61% | aa_slug=gpt-5-6-sol-low · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #9gpt-5-5-highOpenAI | 59.85% | aa_slug=gpt-5-5-high · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #10Claude Opus 4.8Anthropic | 58.33% | aa_slug=claude-opus-4-8 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #11gpt-5-6-terra-highOpenAI | 57.58% | aa_slug=gpt-5-6-terra-high · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #12gpt-5.4OpenAI | 57.58% | aa_slug=gpt-5-4 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #13gpt-5-5-mediumOpenAI | 57.58% | aa_slug=gpt-5-5-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #14gpt-5.6-terraOpenAI | 57.58% | aa_slug=gpt-5-6-terra · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #15claude-opus-4-7-non-reasoningAnthropic | 54.55% | aa_slug=claude-opus-4-7-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #16Gemini 3.1 Pro PreviewGoogle | 53.79% | aa_slug=gemini-3-1-pro-preview · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #17claude-sonnet-4-6-adaptiveAnthropic | 53.03% | aa_slug=claude-sonnet-4-6-adaptive · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #18gpt-5.3-codexOpenAI | 53.03% | aa_slug=gpt-5-3-codex · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #19gpt-5-5-lowOpenAI | 52.27% | aa_slug=gpt-5-5-low · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #20gpt-5.4-miniOpenAI | 52.27% | aa_slug=gpt-5-4-mini · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #21Claude Opus 4.7Anthropic | 51.52% | aa_slug=claude-opus-4-7 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #22Z.ai GLM 5.2Z.ai (Zhipu AI) | 50.76% | aa_slug=glm-5-2 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #23Qwen3.7 MaxQwen | 50.76% | aa_slug=qwen3-7-max · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #24KAT-Coder-Pro V2Kwaipilot | 49.24% | aa_slug=kat-coder-pro-v2 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #25gpt-5-5-non-reasoningOpenAI | 49.24% | aa_slug=gpt-5-5-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #26Claude Opus 4.6Anthropic | 48.48% | aa_slug=claude-opus-4-6 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #27gpt-5.2OpenAI | 46.97% | aa_slug=gpt-5-2 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #28Qwen 3.7 PlusQwen | 46.97% | aa_slug=qwen3-7-plus · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #29claude-opus-4-5-thinkingAnthropic | 46.97% | aa_slug=claude-opus-4-5-thinking · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #30claude-opus-4-6-adaptiveAnthropic | 46.21% | aa_slug=claude-opus-4-6-adaptive · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #31deepseek-v4-pro-0424DeepSeek | 46.21% | aa_slug=deepseek-v4-pro-0424 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #32gemini-3-5-flash-minimalGoogle | 46.21% | aa_slug=gemini-3-5-flash-minimal · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #33Claude Sonnet 4.6Anthropic | 46.21% | aa_slug=claude-sonnet-4-6 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #34gpt-5.1OpenAI | 45.45% | aa_slug=gpt-5-1 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #35muse-sparkMeta AI | 45.45% | aa_slug=muse-spark · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #36Kimi K2.7 CodeMoonshot AI | 44.7% | aa_slug=kimi-k2-7-code · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #37Kimi K2.6Moonshot AI | 43.94% | aa_slug=kimi-k2-6 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #38Qwen3.6 PlusQwen | 43.94% | aa_slug=qwen3-6-plus · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #39gpt-5-6-terra-lowOpenAI | 43.94% | aa_slug=gpt-5-6-terra-low · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #40Qwen3.6 Max PreviewQwen | 43.94% | aa_slug=qwen3-6-max · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #41gpt-5-4-lowOpenAI | 43.18% | aa_slug=gpt-5-4-low · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #42GLM 5.1Z.ai (Zhipu AI) | 43.18% | aa_slug=glm-5-1 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #43MiMo-V2.5-ProXiaomi | 43.18% | aa_slug=mimo-v2-5-pro · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #44GLM 5Z.ai (Zhipu AI) | 43.18% | aa_slug=glm-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #45gpt-5-2-mediumOpenAI | 43.18% | aa_slug=gpt-5-2-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #46gpt-5.4-nanoOpenAI | 42.42% | aa_slug=gpt-5-4-nano · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #47MiniMax M3MiniMax | 42.42% | aa_slug=minimax-m3 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #48claude-sonnet-4-6-non-reasoning-low-effortAnthropic | 42.42% | aa_slug=claude-sonnet-4-6-non-reasoning-low-effort · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #49gpt-5-5-instant-05-26OpenAI | 42.42% | aa_slug=gpt-5-5-instant-05-26 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #50deepseek-v4-pro-0424-highDeepSeek | 41.67% | aa_slug=deepseek-v4-pro-0424-high · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #51MiMo-V2.5Xiaomi | 41.67% | aa_slug=mimo-v2-5-0424 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #52gemini-3-proGoogle | 41.67% | aa_slug=gemini-3-pro · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #53Gemini 3.5 FlashGoogle | 40.91% | aa_slug=gemini-3-5-flash · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #54mimo-v2-proXiaomi | 40.91% | aa_slug=mimo-v2-pro · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #55Claude Opus 4.5Anthropic | 40.91% | aa_slug=claude-opus-4-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #56Qwen3.5 397B A17BQwen | 40.91% | aa_slug=qwen3-5-397b-a17b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #57grok-4-20-0309SpaceXAI | 40.91% | aa_slug=grok-4-20-0309 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #58glm-5-non-reasoningZ.ai (Zhipu AI) | 39.39% | aa_slug=glm-5-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #59gemini-3-5-flash-mediumGoogle | 39.39% | aa_slug=gemini-3-5-flash-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #60MiniMax M2.7MiniMax | 39.39% | aa_slug=minimax-m2-7 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #61deepseek-v4-flash-0420-highDeepSeek | 38.64% | aa_slug=deepseek-v4-flash-0420-high · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #62Gemini 3 Flash PreviewGoogle | 38.64% | aa_slug=gemini-3-flash-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #63grok-4SpaceXAI | 37.88% | aa_slug=grok-4 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #64gpt-5-4-non-reasoningOpenAI | 37.88% | aa_slug=gpt-5-4-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #65Grok 4.3xAI | 37.88% | aa_slug=grok-4-3 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #66GPT-5-CodexOpenAI | 37.88% | aa_slug=gpt-5-codex · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #67Grok 4.20xAI | 37.88% | aa_slug=grok-4-20 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #68kimi-k2-6-non-reasoningMoonshot AI | 37.88% | aa_slug=kimi-k2-6-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #69gpt-5 (medium)OpenAI | 37.88% | aa_slug=gpt-5-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #70GPT-5.2-CodexOpenAI | 37.12% | aa_slug=gpt-5-2-codex · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #71o3OpenAI | 37.12% | aa_slug=o3 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #72Nemotron 3 UltraNVIDIA | 36.36% | aa_slug=nvidia-nemotron-3-ultra-550b-a55b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #73Gemma 4 31BGoogle | 36.36% | aa_slug=gemma-4-31b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #74deepseek-v4-pro-0424-non-reasoningDeepSeek | 36.36% | aa_slug=deepseek-v4-pro-0424-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #75glm-5-1-non-reasoningZ.ai (Zhipu AI) | 35.61% | aa_slug=glm-5-1-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #76mimo-v2-5-pro-non-reasoningXiaomi | 35.61% | aa_slug=mimo-v2-5-pro-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #77DeepSeek V3.2DeepSeek | 35.61% | aa_slug=deepseek-v3-2-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #78claude-4-5-sonnet-thinkingAnthropic | 35.61% | aa_slug=claude-4-5-sonnet-thinking · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #79deepseek-v4-flash-0420DeepSeek | 35.61% | aa_slug=deepseek-v4-flash-0420 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #80Step 3.7 FlashStepFun | 35.61% | aa_slug=step-3-7-flash · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #81qwen3-5-397b-a17b-non-reasoningAlibaba Group | 35.61% | aa_slug=qwen3-5-397b-a17b-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #82mimo-v2-omni-0327Xiaomi | 35.61% | aa_slug=mimo-v2-omni-0327 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #83Qwen3.6 35B A3BQwen | 34.85% | aa_slug=qwen3-6-35b-a3b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #84GPT-5.1-CodexOpenAI | 34.85% | aa_slug=gpt-5-1-codex · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #85MiniMax M2.5MiniMax | 34.85% | aa_slug=minimax-m2-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #86Qwen3.6 27BQwen | 34.85% | aa_slug=qwen3-6-27b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #87Kimi K2.5Moonshot AI | 34.85% | aa_slug=kimi-k2-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #88mimo-v2-omniXiaomi | 34.85% | aa_slug=mimo-v2-omni · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #89deepseek-v3-2-specialeDeepSeek | 34.85% | aa_slug=deepseek-v3-2-speciale · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #90claude-4-1-opus-thinkingAnthropic | 34.34% | aa_slug=claude-4-1-opus-thinking · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #91Hy3 previewTencent | 34.09% | aa_slug=hy3-preview · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #92gemini-3-pro-lowGoogle | 34.09% | aa_slug=gemini-3-pro-low · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #93deepseek-v4-flash-0420-non-reasoningDeepSeek | 34.09% | aa_slug=deepseek-v4-flash-0420-non-reasoning · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #94gpt-5-4-mini-mediumOpenAI | 34.09% | aa_slug=gpt-5-4-mini-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #95gpt-5-4-nano-mediumOpenAI | 33.33% | aa_slug=gpt-5-4-nano-medium · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #96GLM 5 TurboZ.ai (Zhipu AI) | 33.33% | aa_slug=glm-5-turbo · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #97Mistral Medium 3.5Mistral AI | 33.33% | aa_slug=mistral-medium-3-5 · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #98gpt-5-miniOpenAI | 33.33% | aa_slug=gpt-5-mini · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #99GPT-5.1-Codex MiniOpenAI | 33.33% | aa_slug=gpt-5-1-codex-mini · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History | |
| #100Qwen3.5-27BQwen | 32.58% | aa_slug=qwen3-5-27b · variant=hard · evaluator=Artificial Analysis · index_version=4.3 | — (obs. 11 Sept 2026) | artificialanalysis.aiT2 | History |
Scores are reported as published, with their evaluation configuration (harness, prompting, judge). The bar is relative to the best score on this page. Results with different configs are not directly comparable — see methodology.
The config filter matches a value inside each result's configuration (server-side, `config=` on the API). Chips are the values shared by several rows on the first page; per-model identifiers are not offered.
Definition
- Category
- agentic
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Task
- terminal tasks solved by agents
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Metric
- accuracy · %
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
- Direction
- Higher is better
- Website
Source:AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2observed 3 h agomedium
Each value shows its source, tier and observation time. Missing rows mean no source stated them. How results are recorded →
- Evaluated models
- grok-3-mini-reasoning, claude-opus-5-medium, gemma-4-12B, deepseek-v4-flash, grok-4-3-low, GPT-5.1-Codex Mini, Qwen3 Coder Next, KAT-Coder-Pro V2 +523(531 total)
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history
5 claims · 5 properties
Categorycategory1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| agentic | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Metricmetric1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| accuracy | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Tasktask1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| terminal tasks solved by agents | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Unitunit1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| % | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Websitewebsite1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| https://www.tbench.ai | → current | current | AI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2 | medium | curated |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| Artificial Analysis | artificialanalysis.ai/leaderboards/models | leaderboard | T2· Quality secondary | 59 min ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.