Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1815 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Video Understanding | TGIF, MSVD, MSRVTT, ActNet Averaged | 18 | Jun 24, 2026 | ||
| Radio map estimation | SpectrumNet (test) | 18 | May 29, 2026 | ||
| Radio map estimation | UrbanRadio3D (test) | 18 | May 29, 2026 | ||
| Blockchain Anomaly Detection | CryptopiaHacker (test) | 18 | May 29, 2026 | ||
| Code Generation | HumanEval | 18 | May 29, 2026 | ||
| Mathematical Reasoning | GSM8K | 18 | May 29, 2026 | ||
| Question Answering | TriviaQA | 18 | May 29, 2026 | ||
| Image Classification | ImageNet-Sketch | 18 | Jun 18, 2026 | ||
| Table Reasoning | WikiTQ, TabFact, TableBench, HiTab, FinQA Average | 18 | May 29, 2026 | ||
| Table Reasoning | FinQA | 18 | May 29, 2026 | ||
| Table Reasoning | HiTab | 18 | May 29, 2026 | ||
| Table Reasoning | TabFact | 18 | May 29, 2026 | ||
| Table Reasoning | WikiTQ | 18 | May 29, 2026 | ||
| Tool Retrieval | ToolRet CN v1.0 (test) | 18 | May 29, 2026 | ||
| General Reasoning | BigBench-Lite Topic Domains | 18 | May 29, 2026 | ||
| Logical Reasoning | ZebraLogic (held-out) | 18 | May 28, 2026 | ||
| Planning | Tower of Hanoi (held-out) | 18 | May 28, 2026 | ||
| Mathematical Reasoning | MathGAP (held-out) | 18 | May 28, 2026 | ||
| Arithmetic Reasoning | Matchstick Arithmetic (held-out) | 18 | May 28, 2026 | ||
| Medical Question Answering | MedXpertQA-U | 18 | May 28, 2026 | ||
| Medical Question Answering | MedXpertQA-R | 18 | May 28, 2026 | ||
| Image Classification | Dogs & Cats (test) | 18 | May 28, 2026 | ||
| Anomaly Detection and Localization | Real-IAD | 18 | May 28, 2026 | ||
| Embodied Task Completion | ALFWorld ID | 18 | May 28, 2026 | ||
| Web Shopping Agent | WebShop OOD v1.0 | 18 | May 28, 2026 |