Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1261 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Large Language Model Throughput | ShareGPT v3 | 24 | Feb 26, 2026 | ||
| Video Summarization | TVSum | 24 | Feb 26, 2026 | ||
| Own Missing Knowledge (OMK) Prediction | Collaborative Plan Acquisition (CPA) (test) | 24 | Feb 26, 2026 | ||
| Multivariate time series prediction | Electricity | 24 | Feb 26, 2026 | ||
| Fracture Detection | GRAZPEDWRI-DX 1.0 (test) | 24 | Feb 26, 2026 | ||
| Long-form Video Question Answering | EgoSchema | 24 | Mar 16, 2026 | ||
| Commonsense Question Answering | SocialIQA (SIQA) (val) | 24 | Feb 26, 2026 | ||
| Commonsense inference modeling | ATOMIC 20 (test) | 24 | Feb 26, 2026 | ||
| Highlight Detection | TVSum v1.0 (test) | 24 | Feb 26, 2026 | ||
| Model Inversion | CelebA (private) | 24 | Feb 26, 2026 | ||
| Membership Inference Attack | WikiMIA length 256 | 24 | Feb 26, 2026 | ||
| Membership Inference Attack | WikiMIA | 24 | Feb 26, 2026 | ||
| Novel view synthesis | D-NeRF Hellwarrior | 24 | May 12, 2026 | ||
| Novel view synthesis | D-NeRF Trex | 24 | May 12, 2026 | ||
| Dichotomous Image Segmentation | DIS5K DIS-TE Overall (test) | 24 | May 7, 2026 | ||
| Dichotomous Image Segmentation | DIS5K DIS-TE4 (test) | 24 | May 15, 2026 | ||
| Dichotomous Image Segmentation | DIS5K DIS-TE3 (test) | 24 | May 15, 2026 | ||
| Dichotomous Image Segmentation | DIS5K DIS-TE2 (test) | 24 | May 15, 2026 | ||
| Dichotomous Image Segmentation | DIS5K DIS-TE1 (test) | 24 | May 15, 2026 | ||
| Image Quality Assessment | nuScenes nighttime (val) | 24 | Feb 26, 2026 | ||
| Action Segmentation | 50 Salads (eval) | 24 | Jul 7, 2026 | ||
| Clustering (Color) | Flowers | 24 | Feb 26, 2026 | ||
| Clustering (Color) | Stanford Cars | 24 | Feb 26, 2026 | ||
| Clustering (Pose) | CMUface | 24 | Feb 26, 2026 | ||
| Clustering (Glass) | CMUface | 24 | Feb 26, 2026 |