Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 760 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Stage-aware Prefill | TULU-3 (dev) | 32 | May 8, 2026 | ||
| Semantic Segmentation | RAW ADE20K (Over-exposure (OE)) | 32 | May 11, 2026 | ||
| Semantic Segmentation | ADE20K RAW Normal (NM) | 32 | May 11, 2026 | ||
| Popular Path Search | Melb | 32 | May 8, 2026 | ||
| Popular Path Search | Disland | 32 | May 8, 2026 | ||
| Popular Path Search | Toro | 32 | May 8, 2026 | ||
| Popular Path Search | Edin | 32 | May 8, 2026 | ||
| Popular Path Search | CaliAdv | 32 | May 8, 2026 | ||
| Popular Path Search | Epcot | 32 | May 8, 2026 | ||
| Popular Path Search | DisHolly | 32 | May 8, 2026 | ||
| Faithfulness evaluation of image explanation | ImageNet (val) | 32 | May 8, 2026 | ||
| Image Regression | RSRC-L8 (test) | 32 | May 8, 2026 | ||
| Image Regression | RSRC-S2 (test) | 32 | May 8, 2026 | ||
| Mathematical Reasoning | AIME | 32 | Jun 25, 2026 | ||
| Safety Evaluation | SorryBench | 32 | May 7, 2026 | ||
| Classification | Epsilon (test) | 32 | Jun 19, 2026 | ||
| Unlearning | Ele-Fashion | 32 | May 6, 2026 | ||
| Text-to-SQL | Spider Non-Synthesized Matched Set | 32 | May 1, 2026 | ||
| Text-to-SQL | BIRD (Non-Synthesized Matched Set) | 32 | May 1, 2026 | ||
| Text-to-SQL | BIRD Synthesized Matched Set | 32 | May 1, 2026 | ||
| Multi-turn jailbreak defense | Multi-turn Jailbreak Defense | 32 | May 1, 2026 | ||
| Image Classification | Oxford Flowers 102 | 32 | Jun 23, 2026 | ||
| Classification | Forest Type Mapping Standard (5-fold CV) | 32 | May 1, 2026 | ||
| Continual Category Discovery | Average fine-grained | 32 | May 1, 2026 | ||
| Out-of-Distribution Detection | ImageNet-1k OOD Average (val) | 32 | May 25, 2026 |