Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1078 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Jailbreak Attack | FigStep | 26 | Apr 15, 2026 | ||
| Machine Translation | WMT de-en & fr-en 14 (test) | 26 | Feb 26, 2026 | ||
| Summarization | XSUM & AESLC WMT-14 (test) | 26 | Feb 26, 2026 | ||
| Planning | nuScenes | 26 | Jul 8, 2026 | ||
| Long-context language understanding | L-Eval | 26 | Feb 26, 2026 | ||
| Long-context language understanding | L-Eval (test) | 26 | Feb 26, 2026 | ||
| Image Classification | ImageNet-1k 1.0 (val) | 26 | Feb 26, 2026 | ||
| Node Classification | Tolokers v1 (test) | 26 | Feb 26, 2026 | ||
| Code Reasoning | CRUX | 26 | May 12, 2026 | ||
| 3D Talking Head Generation | DualTalk OOD set | 26 | Feb 26, 2026 | ||
| Fact-Checking | PubHealth | 26 | Feb 26, 2026 | ||
| Gait Recognition | CASIA-B | 26 | May 22, 2026 | ||
| Language Modeling | FW (val) | 26 | Feb 26, 2026 | ||
| Process-level verification | MATH ProcessBench (test) | 26 | Feb 26, 2026 | ||
| Autonomous Driving | Bench2Drive base subset (50 clips for open-loop) | 26 | May 22, 2026 | ||
| Mathematical problem solving | MATH-Vision (test) | 26 | Feb 26, 2026 | ||
| Person Re-identification | PRCC Clothes-Changing (CC) (test) | 26 | Apr 23, 2026 | ||
| Stereo Matching | Booster Q (test) | 26 | Feb 26, 2026 | ||
| Machine-Generated Text Detection | SemEval (test) | 26 | Feb 26, 2026 | ||
| Image Restoration | Realsr | 26 | Mar 24, 2026 | ||
| 3D Semantic Occupancy Prediction | OpenOccupancy | 26 | Jun 23, 2026 | ||
| Referring Video Object Segmentation | ReVOS | 26 | Jun 25, 2026 | ||
| Visual Question Answering | Ultra-MedVQA Task 6 | 26 | Feb 26, 2026 | ||
| Visual Question Answering | Ultra-MedVQA Task 5 | 26 | Feb 26, 2026 | ||
| Visual Question Answering | Ultra-MedVQA Task 4 | 26 | Feb 26, 2026 |