Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1995 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Health Dialogue | HealthBench | 17 | Jun 2, 2026 | ||
| Open-ended Writing | ArenaHard | 17 | Jun 2, 2026 | ||
| Violence Detection | RWF-2000 (test) | 17 | Jul 7, 2026 | ||
| Robot Manipulation | Robosuite Can | 17 | Jun 2, 2026 | ||
| Robot Manipulation | Robosuite Lift | 17 | Jun 2, 2026 | ||
| Generalization Bound Estimation | ImageNet (val) | 17 | Jun 2, 2026 | ||
| Time-series forecasting | LOB | 17 | Jun 2, 2026 | ||
| Time-series forecasting | Crypto | 17 | Jun 2, 2026 | ||
| System-level data generation | Stock | 17 | Jun 2, 2026 | ||
| System-level data generation | LOB | 17 | Jun 2, 2026 | ||
| Temperature Reconstruction | Subcooled Pool Boiling Task 1 (50 random snapshots) | 17 | Jun 2, 2026 | ||
| Compositional Retrieval | ScanNet Apples-to-apples intersection (original & expansion compounds) | 17 | Jun 2, 2026 | ||
| Image Understanding | SEED-IMG | 17 | Jun 2, 2026 | ||
| Machine Unlearning | TOFU | 17 | Jun 1, 2026 | ||
| Egocentric Video Understanding | EgoSchema Subset 2023 | 17 | Jun 1, 2026 | ||
| 3D Object Detection | nuScenes Camera View Drop 6 drops | 17 | Jun 1, 2026 | ||
| 3D Object Detection | nuScenes LiDAR Object Failure rate = 0.5 | 17 | Jun 1, 2026 | ||
| 3D Object Detection | nuScenes LiDAR Limited FOV [-60, 60] | 17 | Jun 1, 2026 | ||
| 3D Object Detection | nuScenes LiDAR Drop all | 17 | Jun 1, 2026 | ||
| Object Counting | CLOC (test) | 17 | Jun 1, 2026 | ||
| Locomotion | D4RL walker2d-medium-expert v2 | 17 | Jun 16, 2026 | ||
| Mechanical Drawing Understanding | MechVQA (test) | 17 | Jun 1, 2026 | ||
| Knowledge Graph Completion | UMLS (test) | 17 | Jun 1, 2026 | ||
| Knowledge Graph Completion | Family (test) | 17 | Jun 1, 2026 | ||
| Interpolation for continuous time dynamics | Macrophage Stimulus PCSK3 (test) | 17 | Jun 1, 2026 |