Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1506 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Anomaly Detection | OrganA (test) | 21 | Feb 26, 2026 | ||
| Frame Retrieval | Tennis Forehand | 21 | Feb 26, 2026 | ||
| Action Phase Classification | Tennis Forehand | 21 | Feb 26, 2026 | ||
| Frame Retrieval | Pour Liquid | 21 | Feb 26, 2026 | ||
| Action Phase Classification | Pour Liquid | 21 | Feb 26, 2026 | ||
| Frame Retrieval | Pour Milk | 21 | Feb 26, 2026 | ||
| Action Phase Classification | Pour Milk | 21 | Feb 26, 2026 | ||
| Capacitated Vehicle Routing Problem | CVRP n=100 (10k instances) | 21 | Feb 26, 2026 | ||
| Sequential Recommendation | Amazon Sports and Outdoors | 21 | Feb 26, 2026 | ||
| Node classification | Synthetic homo ratio 0.9 | 21 | Feb 26, 2026 | ||
| Node classification | Synthetic homo ratio 0.8 | 21 | Feb 26, 2026 | ||
| Node classification | Synthetic homo ratio 0.5 | 21 | Feb 26, 2026 | ||
| Image Generation | LSUN-Bedroom 50k samples (test) | 21 | Feb 26, 2026 | ||
| Performance Difference Estimation | Adult Subgroup #5, 1% | 21 | Feb 26, 2026 | ||
| Performance Difference Estimation | Adult Subgroup #4, 1% | 21 | Feb 26, 2026 | ||
| Performance Difference Estimation | Adult Subgroup #3, 3% | 21 | Feb 26, 2026 | ||
| Performance Difference Estimation | Adult Subgroup #2, 9% | 21 | Feb 26, 2026 | ||
| Performance Difference Estimation | Adult Subgroup #1, 86% | 21 | Feb 26, 2026 | ||
| Multimodal Classification | UR-FUNNY | 21 | Apr 6, 2026 | ||
| Question Answering | SQuAD v1.1 & v2.0 | 21 | Mar 10, 2026 | ||
| Visual Reasoning | NLVR2 (test-P) | 21 | Feb 26, 2026 | ||
| Question Answering | TabMWP (test) | 21 | Feb 26, 2026 | ||
| Molecule Generation | Isocyanates | 21 | Jun 4, 2026 | ||
| Fine-grained Image Classification | Herbarium (test) | 21 | Feb 26, 2026 | ||
| Fine-grained Image Classification | EuroSAT (test) | 21 | Feb 26, 2026 |