Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2015 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Video Understanding | Opencompass Video Benchmark | 17 | Apr 14, 2026 | ||
| Agent Interaction | Aggregate FTWP, ScienceWorld, WebShop | 17 | Apr 14, 2026 | ||
| Agent Interaction | ScienceWorld (test) | 17 | Apr 14, 2026 | ||
| Agent Interaction | ScienceWorld (val) | 17 | Apr 14, 2026 | ||
| Agent Interaction | FTWP (test) | 17 | Apr 14, 2026 | ||
| Agent Interaction | FTWP (val) | 17 | Apr 14, 2026 | ||
| Controllable writing | WritingBench (WB) | 17 | Apr 14, 2026 | ||
| Open-ended generation | HelloBench (HB) | 17 | Apr 14, 2026 | ||
| Question Answering | TriviaQA | 17 | Apr 14, 2026 | ||
| Question Answering | PopQA | 17 | Apr 14, 2026 | ||
| Legal Question Answering | LawBench revised (test) | 17 | Apr 14, 2026 | ||
| Low-light image enhancement | LOL syn v2 | 17 | Apr 14, 2026 | ||
| Multi-view Composed Image Retrieval | Multi-dataset Average (DeepFashion, F200K, FashionGen) (val) | 17 | Apr 14, 2026 | ||
| Multi-view Composed Image Retrieval | FashionGen (val) | 17 | Apr 14, 2026 | ||
| Multi-view Composed Image Retrieval | F200K (val) | 17 | Apr 14, 2026 | ||
| Multi-view Composed Image Retrieval | DeepFashion (val) | 17 | Apr 14, 2026 | ||
| Inductive Temporal Knowledge Graph Link Prediction | GDELT (5:2:3 chronological split) | 17 | Apr 14, 2026 | ||
| Inductive Temporal Knowledge Graph Link Prediction | ICEWS 05-15 (5:2:3 chronological split) | 17 | Apr 14, 2026 | ||
| Inductive Temporal Knowledge Graph Link Prediction | ICEWS 18 (5:2:3 chronological split) | 17 | Apr 14, 2026 | ||
| Inductive Temporal Knowledge Graph Link Prediction | ICEWS 14 (5:2:3 chronological split) | 17 | Apr 14, 2026 | ||
| MDD Classification | REST-meta-MDD (10-fold Cross-val) | 17 | Apr 14, 2026 | ||
| Social Reasoning | Sotopia hard | 17 | Apr 14, 2026 | ||
| Code Generation | Code Benchmarks (HumanEval, MBPP) | 17 | Apr 14, 2026 | ||
| Tool Calling | DIABENCH Dynamic Evaluation 1.0 | 17 | Apr 14, 2026 | ||
| Tool Calling | DIABENCH Static Evaluation 1.0 | 17 | Apr 14, 2026 |