Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 835 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Commonsense Reasoning | CSQA | 30 | May 28, 2026 | ||
| General Knowledge Question Answering | MMLU | 30 | May 28, 2026 | ||
| Web Browsing | BC-plus | 30 | May 28, 2026 | ||
| Multi-hop Question Answering | SQuAD | 30 | May 28, 2026 | ||
| Probabilistic Forecasting | Metaculus and Polymarket (test) | 30 | May 28, 2026 | ||
| Causal Discovery | Auto | 30 | May 27, 2026 | ||
| Preference Modeling | HH-RLHF | 30 | May 27, 2026 | ||
| Reasoning | DeepScaler | 30 | May 27, 2026 | ||
| Source Attribution Detection | WikiMIA source-attribution | 30 | May 27, 2026 | ||
| Sparse GEMM | Matrix Dimensions (M, N, K) H20 GPU | 30 | May 27, 2026 | ||
| Mean Reward | WebShop | 30 | May 27, 2026 | ||
| Mean Reward | ScienceWorld | 30 | May 27, 2026 | ||
| Mean Reward | ALFWorld | 30 | May 27, 2026 | ||
| Code Generation | MBPP | 30 | Jun 11, 2026 | ||
| Visual Reasoning | Commonsense Reasoning | 30 | May 26, 2026 | ||
| Visual Reasoning | Quantitative Reasoning | 30 | May 26, 2026 | ||
| Visual Reasoning | Physical Reasoning | 30 | May 26, 2026 | ||
| Camouflaged Object Detection | CAMO (250) | 30 | May 26, 2026 | ||
| Sorting | Synthetic Sorting Value range [10, 11] | 30 | May 26, 2026 | ||
| Sorting | Synthetic Sorting Value range [0, 1] | 30 | May 26, 2026 | ||
| Mathematical Reasoning | Minerva | 30 | May 26, 2026 | ||
| Sequential Decision Making | RockSample (8, 4, 10, -1) | 30 | May 26, 2026 | ||
| Sequential Decision Making | RockSample (7, 8, 20, 0) | 30 | May 26, 2026 | ||
| Sequential Decision Making | Tag | 30 | May 26, 2026 | ||
| Image Reconstruction | QPI | 30 | May 26, 2026 |