Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2071 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Generation | ImageNet-1k 1.0 (val) | 17 | Apr 14, 2026 | ||
| Molecule Graph Generation | GuacaMol (test) | 17 | Feb 26, 2026 | ||
| Referring Image Segmentation | RefCOCOg UMD (val) | 17 | Feb 26, 2026 | ||
| Underwater Image Enhancement | LSUI L400 (test) | 17 | Mar 25, 2026 | ||
| Multivariate long-term forecasting | Traffic standard (test) | 17 | Jun 11, 2026 | ||
| Camera tracking | TUM dynamic scene sequences RGB-D (test) | 17 | Feb 26, 2026 | ||
| Multimodal Emotion Recognition | DFEW | 17 | Jun 9, 2026 | ||
| Code Generation | xCodeEval (sample 500) | 17 | Feb 26, 2026 | ||
| Text-to-Motion | Motion-X | 17 | Feb 27, 2026 | ||
| Traffic Signal Control | D_HZ | 17 | Feb 26, 2026 | ||
| Long-context Question Answering | Qasper | 17 | Feb 27, 2026 | ||
| Continual Learning | CUB | 17 | Mar 12, 2026 | ||
| Monocular Depth Estimation | DIODE (test) | 17 | Mar 13, 2026 | ||
| Multiple-Choice Video QA | EgoSchema latest (test) | 17 | Feb 26, 2026 | ||
| Image Classification | CMNIST 0.5% bias ratio unbiased (test) | 17 | Feb 26, 2026 | ||
| Unconditional Image Generation | CIFAR-10 unconditional 32x32 | 17 | Feb 26, 2026 | ||
| Referring Expression Comprehension | RefCOCOg 48 (test-u) | 17 | Feb 26, 2026 | ||
| Referring Expression Comprehension | RefCOCOg 48 (val-u) | 17 | Feb 26, 2026 | ||
| Referring Expression Comprehension | RefCOCO+ 80 (val) | 17 | Feb 26, 2026 | ||
| Referring Expression Comprehension | RefCOCO 80 (testB) | 17 | Feb 26, 2026 | ||
| Referring Expression Comprehension | RefCOCO 80 (testA) | 17 | Feb 26, 2026 | ||
| Referring Expression Comprehension | RefCOCO 80 (val) | 17 | Feb 26, 2026 | ||
| Task Planning | Hugging Face v1 (test) | 17 | Feb 26, 2026 | ||
| 3D Object Detection | KITTI | 17 | May 19, 2026 | ||
| Partial Label Learning | CIFAR-100 | 17 | Mar 4, 2026 |