Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 399 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Diffusion Adaptation | Barbie | 48 | Jul 1, 2026 | ||
| Long-context Recall | Long-context Recall-intensive Tasks Suite SQuAD, FDA, SWDE | 48 | Jun 30, 2026 | ||
| Text-to-image retrieval | IMDB | 48 | Jun 30, 2026 | ||
| Multi-task Image Classification | 20 Vision Tasks | 48 | Jun 26, 2026 | ||
| Multi-task Image Classification | 14 vision tasks | 48 | Jun 26, 2026 | ||
| Multi-task Image Classification | 8 vision tasks | 48 | Jun 26, 2026 | ||
| Image Reconstruction | RxRx1 and SR-Caco-2 (test) | 48 | Jun 25, 2026 | ||
| Action Recognition | Ego-Exo4D | 48 | Jun 25, 2026 | ||
| Visual Place Recognition | XHZ | 48 | Jun 24, 2026 | ||
| Graph Alignment | CoraFull (test) | 48 | Jun 24, 2026 | ||
| Graph Alignment | OGBN-Arxiv (test) | 48 | Jun 24, 2026 | ||
| Graph Alignment | PCQM4M v2 (test) | 48 | Jun 24, 2026 | ||
| Graph Alignment | AQSOL (test) | 48 | Jun 24, 2026 | ||
| Graph Alignment | Erdös-Rényi (test) | 48 | Jun 24, 2026 | ||
| Human agreement correlation | VideoPhy-2 (test) | 48 | Jun 23, 2026 | ||
| Dueling Bandits | Condorcet-winner dueling bandits random instances | 48 | Jun 23, 2026 | ||
| Image Reconstruction | STL (test) | 48 | Jun 23, 2026 | ||
| Competition Mathematical Reasoning | MATH-500 | 48 | Jun 23, 2026 | ||
| Evidence-round recall | LongMemEval 100 question sample | 48 | Jun 19, 2026 | ||
| Distribution Alignment | ImageNet (val) | 48 | Jun 18, 2026 | ||
| Distribution Alignment | ImageNet (train) | 48 | Jun 18, 2026 | ||
| Generative Modeling | ImageNet (val) | 48 | Jun 18, 2026 | ||
| Watermark Extraction | Math domain dataset | 48 | Jun 17, 2026 | ||
| Watermark Extraction | Legal domain dataset | 48 | Jun 17, 2026 | ||
| Watermark Extraction | Medical domain dataset | 48 | Jun 17, 2026 |