Public Artificial Intelligence Benchmark Records Database
Benchmarks
Table may be scrolled horizontally on narrow displays.
| Benchmark | Version | Organization | Status | Records |
|---|---|---|---|---|
| AA-LCR | Unknown | Artificial Analysis | ACTIVE | 371 |
| AA-Omniscience | Unknown | Artificial Analysis | ACTIVE | 702 |
| AIME | Unknown | MAA (Mathematical Association of America) | ACTIVE | 3 |
| AIRS-Bench | Unknown | Meta FAIR | ACTIVE | 2 |
| Apex | Unknown | Mercor | ACTIVE | 2 |
| Apex Shortlist | Unknown | Unknown | ACTIVE | 2 |
| APEX-Agents | Unknown | Mercor | ACTIVE | 20 |
| ARC-AGI | Unknown | ARC Prize Foundation | ACTIVE | 1 |
| Artificial Analysis Intelligence Index | Unknown | Artificial Analysis | ACTIVE | 398 |
| Automation Bench | Unknown | Zapier | ACTIVE | 2 |
| BFCL-V4 | v4 | UC Berkeley | ACTIVE | 1 |
| BrowseComp | Unknown | OpenAI | ACTIVE | 9 |
| BrowseComp-zh | Unknown | Unknown | ACTIVE | 1 |
| C-Eval | Unknown | Unknown | ACTIVE | 1 |
| ChartQAPro | Unknown | York University | ACTIVE | 2 |
| Chinese-SimpleQA | Unknown | Alibaba | ACTIVE | 2 |
| Claw-Eval | Unknown | Unknown | ACTIVE | 1 |
| Codeforces | Unknown | Codeforces | ACTIVE | 2 |
| CorpusQA | Unknown | Alibaba | ACTIVE | 2 |
| CritPt | Unknown | Unknown | ACTIVE | 355 |
| CursorBench | Unknown | Cursor (Anysphere) | ACTIVE | 1 |
| DeepPlanning | Unknown | Alibaba (Qwen Team) | ACTIVE | 1 |
| DeepSWE | Unknown | Datacurve | ACTIVE | 15 |
| Finance Agent | Unknown | Vals AI | ACTIVE | 2 |
| FrontierSWE | Unknown | Proximal AI | ACTIVE | 1 |
| GDPval-AA | v2 | Artificial Analysis | ACTIVE | 129 |
| GPQA | Unknown | New York University | ACTIVE | 406 |
| GraphWalks | Unknown | OpenAI | ACTIVE | 4 |
| HLE-Verified | Unknown | Alibaba (Qwen Team) | ACTIVE | 1 |
| HMMT | Unknown | Harvard-MIT Mathematics Tournament | ACTIVE | 8 |
| Humanity's Last Exam | Unknown | Center for AI Safety | ACTIVE | 411 |
| IFBench | Unknown | Allen Institute for AI | ACTIVE | 358 |
| IFEval | Unknown | ACTIVE | 1 | |
| IMOAnswerBench | Unknown | Google DeepMind | ACTIVE | 5 |
| ITBench SRE | Unknown | IBM | ACTIVE | 23 |
| LiveCodeBench | Unknown | UC Berkeley | ACTIVE | 2 |
| LiveCodeBench Pro | Unknown | Unknown | ACTIVE | 1 |
| LiveCodeBench v6 | v6 | UC Berkeley | ACTIVE | 1 |
| LongBench v2 | v2 | Tsinghua University | ACTIVE | 1 |
| MCP Atlas | Unknown | Scale AI | ACTIVE | 7 |
| MCP-Mark | Unknown | Unknown | ACTIVE | 1 |
| MMLU-Pro | Unknown | University of Waterloo (TIGER-Lab) | ACTIVE | 2 |
| MMMLU | Unknown | OpenAI | ACTIVE | 3 |
| MMMU-Pro | Unknown | Unknown | ACTIVE | 153 |
| MRCR | Unknown | Google DeepMind | ACTIVE | 2 |
| MRCR v2 | v2 | Google DeepMind | ACTIVE | 2 |
| MultiChallenge | Unknown | Scale AI | ACTIVE | 1 |
| NL2Repo | Unknown | Unknown | ACTIVE | 2 |
| OSWorld-Verified | Unknown | XLang Lab (University of Hong Kong) | ACTIVE | 2 |
| PostTrainBench | Unknown | Thoughtful | ACTIVE | 1 |
| ProgramBench | Unknown | Meta FAIR | ACTIVE | 1 |
| SciCode | Unknown | University of Illinois Urbana-Champaign | ACTIVE | 402 |
| ScreenSpot-Pro | Unknown | National University of Singapore | ACTIVE | 2 |
| Seal-0 | Unknown | Virginia Tech | ACTIVE | 1 |
| SecCodeBench | Unknown | Alibaba | ACTIVE | 1 |
| SimpleQA-Verified | Unknown | Google DeepMind | ACTIVE | 2 |
| SkillsBench | Unknown | BenchFlow AI | ACTIVE | 1 |
| SuperGPQA | Unknown | ByteDance | ACTIVE | 1 |
| SWE-bench | Unknown | Princeton University | ACTIVE | 25 |
| SWE-Marathon | Unknown | Abundant AI | ACTIVE | 1 |
| tau2-bench | Unknown | Sierra | ACTIVE | 353 |
| tau3-Banking | Unknown | Artificial Analysis | ACTIVE | 119 |
| Terminal-Bench | Unknown | Laude Institute | ACTIVE | 121 |
| Terminal-Bench Hard | Unknown | Laude Institute | ACTIVE | 345 |
| Tool Decathlon | Unknown | HKUST | ACTIVE | 2 |
| Toolathlon | Unknown | HKUST | ACTIVE | 2 |
| Video-MME | Unknown | Tencent | ACTIVE | 1 |
| VITA-Bench | Unknown | Meituan | ACTIVE | 1 |
| WideSearch | Unknown | ByteDance | ACTIVE | 1 |