Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1252 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Harmful Request Defense | SORRY-Bench | 24 | Feb 26, 2026 | ||
| Competition Mathematics | AIME 2025 | 24 | Feb 26, 2026 | ||
| Competition Mathematics | AIME 2024 | 24 | Feb 26, 2026 | ||
| Strategic Reasoning | CheckmateInOne | 24 | Feb 26, 2026 | ||
| Algorithmic Reasoning | Word Sorting | 24 | Feb 26, 2026 | ||
| Logical Reasoning | AR-LSAT (test) | 24 | Feb 26, 2026 | ||
| Permutation Flow Shop Scheduling Problem | PFSP | 24 | May 11, 2026 | ||
| Multi-hop Question Answering | Average (2WikiMQA, Bamboogle, Frames, MuSiQue) | 24 | Feb 26, 2026 | ||
| Multi-hop Question Answering | MuSiQue | 24 | Feb 26, 2026 | ||
| Multi-hop Question Answering | Frames | 24 | Feb 26, 2026 | ||
| Tool-use | BFCL Multi-turn | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | HMMT February | 24 | Apr 21, 2026 | ||
| Overall Performance | LoCoMo | 24 | May 14, 2026 | ||
| Opinion Forecasting | Delhi Election (test) | 24 | Feb 26, 2026 | ||
| Commonsense Reasoning | Commonsense Reasoning (OBQA, ARC-C, Wino, PIQA, Social, ARC-E, BoolQ, Hella) | 24 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 25 | 24 | Feb 26, 2026 | ||
| Comparative Performance Evaluation | DeepConsult | 24 | Feb 26, 2026 | ||
| Toxicity Detection | Hosseini’s 1st Negated Sentence 1.0 (test) | 24 | Feb 26, 2026 | ||
| Fermi Problem Solving | Fermi | 24 | Feb 26, 2026 | ||
| Imputation | Sleep EDF Drop 6 Channel | 24 | Feb 26, 2026 | ||
| Imputation | Sleep EDF Drop 3 Channel | 24 | Feb 26, 2026 | ||
| Imputation | Sleep EDF Drop 1 Channel | 24 | Feb 26, 2026 | ||
| Text-to-Image Generation | SDXL 200 prompts (val) | 24 | Feb 26, 2026 | ||
| Video Question Answering | Next-QA v1 (test) | 24 | Feb 26, 2026 | ||
| Video-to-Text Retrieval | MSRVTT | 24 | Feb 26, 2026 |