Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1571 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Speculative Decoding | 20 Prompts across 4 Task Categories | 20 | May 5, 2026 | ||
| Image Classification | Vision Classification Suite (CIFAR-100, EuroSAT, RESISC45, Stanford Cars, FGVC Aircraft, DTD, CIFAR-10, Oxford-IIIT Pets) (test) | 20 | May 5, 2026 | ||
| Image Deraining | Rain100H | 20 | Jun 23, 2026 | ||
| Diagram Understanding | AI2D lite | 20 | May 4, 2026 | ||
| Multimodal Understanding | MMT emo | 20 | May 4, 2026 | ||
| Multimodal Understanding | MMBench-EN lite | 20 | May 4, 2026 | ||
| Multimodal Understanding | MMBench-CN lite | 20 | May 4, 2026 | ||
| Molecular regression | ZINC small 12K | 20 | May 4, 2026 | ||
| Super-resolution | Sentinel-2 (val) | 20 | May 4, 2026 | ||
| OOD Detection | CURE-OOD OS Exposure Time shift | 20 | May 4, 2026 | ||
| Multi-turn Dialogue | MT-Bench | 20 | May 4, 2026 | ||
| Protein variant effect prediction | ProteinGym substitution benchmark | 20 | May 28, 2026 | ||
| Tool-use | Evaluation dataset | 20 | May 4, 2026 | ||
| Document Question Answering | DocVQA | 20 | Jun 19, 2026 | ||
| Visual Question Answering | SpecVQA | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MTWQ in-target | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MRUC in-target | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MWTWT in-target | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MCCQ in-target | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MTWQ | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MRUC | 20 | May 1, 2026 | ||
| Multi-modal Stance Detection | MWTWT | 20 | May 1, 2026 | ||
| Caption Evaluation | nocaps | 20 | May 1, 2026 | ||
| Caption Evaluation | MSCOCO | 20 | May 1, 2026 | ||
| Fact Verification | Fact Verification | 20 | May 1, 2026 |