Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1299 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Cultural Reasoning | FORK | 23 | Apr 21, 2026 | ||
| Multilingual Math Reasoning | MT-AIME | 23 | Apr 21, 2026 | ||
| Multilingual Math Reasoning | MGSM | 23 | Apr 21, 2026 | ||
| Image Classification | CIFAR-10 | 23 | Apr 21, 2026 | ||
| Long-term time series forecasting | Weather Average | 23 | May 19, 2026 | ||
| Camouflaged Object Detection | Weeds-Banana (test) | 23 | Apr 20, 2026 | ||
| 3D Object Classification | ScanObjectNN ONLY | 23 | Apr 20, 2026 | ||
| Math | AIME 25 | 23 | Jun 2, 2026 | ||
| Chart Visual Question Answering | CHARTOM non-misleading visualizations | 23 | Apr 20, 2026 | ||
| Chart Visual Question Answering | CHARTOM misleading visualizations | 23 | Apr 20, 2026 | ||
| Human-Object Interaction Detection | HICO-DET computed-box setting | 23 | Apr 16, 2026 | ||
| Information Retrieval | Korean Web Search Offline Evaluation set (test) | 23 | Apr 16, 2026 | ||
| Instruction Following | MultiChallenge (Out-of-Domain) | 23 | Apr 15, 2026 | ||
| Instruction Following | AgentIF (Out-of-Domain) | 23 | Apr 15, 2026 | ||
| Instruction Following | Collie (Out-of-Domain) | 23 | Apr 15, 2026 | ||
| Instruction Following | WritingBench (Out-of-Domain) | 23 | Apr 15, 2026 | ||
| Instruction Following | ComplexBench Out-of-Domain | 23 | Apr 15, 2026 | ||
| Instruction Following | FollowBench In-Domain | 23 | Apr 15, 2026 | ||
| Instruction Following | CFBench In-Domain | 23 | Apr 15, 2026 | ||
| Instruction Following | IFEval In-Domain | 23 | Apr 15, 2026 | ||
| Language Modeling | OWT (val) | 23 | Jun 4, 2026 | ||
| Empathic Dialogue | Lend-an-Ear (test) | 23 | Apr 14, 2026 | ||
| Node Classification | Squirrel fix (test) | 23 | Apr 14, 2026 | ||
| Retrieval-Augmented Question Answering | DeepSearch Average | 23 | Apr 14, 2026 | ||
| Retrieval-Augmented Question Answering | DeepSearch Bamboogle | 23 | Apr 14, 2026 |