Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1818 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Adversarial Attack | MS-COCO (test) | 18 | May 27, 2026 | ||
| Safety Alignment Evaluation | HEx-PHI | 18 | May 27, 2026 | ||
| Safety Alignment Evaluation | SorryBench | 18 | May 27, 2026 | ||
| Jailbreak Robustness | Jailbreak Attacks | 18 | May 27, 2026 | ||
| Natural Language Understanding | EuroEval French | 18 | May 27, 2026 | ||
| Natural Language Understanding | EuroEval Spanish | 18 | May 27, 2026 | ||
| Natural Language Understanding | EuroEval English | 18 | May 27, 2026 | ||
| Natural Language Understanding | EuroEval German | 18 | May 27, 2026 | ||
| Visual Object Tracking | COESOT | 18 | May 27, 2026 | ||
| Image Classification | STL10 | 18 | May 27, 2026 | ||
| Image Classification | Flowers102 | 18 | May 27, 2026 | ||
| Image Classification | DTD | 18 | May 27, 2026 | ||
| OOD Generalization | VLCS | 18 | May 27, 2026 | ||
| Answering queries with two free variables | FB15k | 18 | May 26, 2026 | ||
| Multimodal Evaluation | MME | 18 | May 26, 2026 | ||
| Legal Examination | legal examination tasks Out-of-domain | 18 | May 26, 2026 | ||
| Legal Research and Reasoning | LegalSearch In-domain | 18 | May 26, 2026 | ||
| Merge Conflict Resolution | Merge-Bench Java (test) | 18 | May 26, 2026 | ||
| Capacitated Vehicle Routing Problem with Time Windows | CVRPTW100 1,000 instances | 18 | Jun 2, 2026 | ||
| Graph Classification | Computers | 18 | May 26, 2026 | ||
| Graph Classification | CiteSeer | 18 | May 26, 2026 | ||
| Research Automation | DRB | 18 | May 26, 2026 | ||
| Research Automation | MAB.research | 18 | May 26, 2026 | ||
| Code Generation | MAB code | 18 | May 26, 2026 | ||
| Bounding box classification | LVIS 1203 | 18 | May 26, 2026 |