Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2003 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Clustering | synthetic | 17 | May 15, 2026 | ||
| Clustering | synthetic | 17 | May 15, 2026 | ||
| Offline-to-Online Reinforcement Learning | D4RL Aggregate | 17 | Jul 2, 2026 | ||
| OOD Detection | CIFAR-10 OpenOOD v1.5 (test) | 17 | May 15, 2026 | ||
| OOD Detection | ImageNet OOD Aggregate Average | 17 | May 15, 2026 | ||
| Point cloud reconstruction | N-RGBD | 17 | Jun 18, 2026 | ||
| Vocabulary difficulty prediction | KVL data (test) | 17 | May 15, 2026 | ||
| Few-shot classification | CDD 5-way 5-shot | 17 | May 15, 2026 | ||
| Confidence Estimation (Iterative Tagging) | PopQA | 17 | May 15, 2026 | ||
| Confidence Estimation (Iterative Tagging) | Bios | 17 | May 15, 2026 | ||
| Language Modeling | Evaluation Tasks Zero-shot Average | 17 | May 15, 2026 | ||
| Image Question Answering | MMBench | 17 | May 14, 2026 | ||
| Class-Incremental Learning | FGVC-Aircraft B0 Inc10 (test) | 17 | May 28, 2026 | ||
| Goal-reaching | AntMaze Teleport (navigate-v0) | 17 | Jun 4, 2026 | ||
| Action Recognition | Kinetics (meta-test) | 17 | May 14, 2026 | ||
| Reasoning | Xcope bo | 17 | May 14, 2026 | ||
| Multiple-choice Reasoning | Hellaswag bo | 17 | May 14, 2026 | ||
| Link Prediction | Cora | 17 | May 14, 2026 | ||
| Node Classification | Cora | 17 | May 14, 2026 | ||
| Average Performance | CiteSeer | 17 | May 14, 2026 | ||
| Average Performance | Cora | 17 | May 14, 2026 | ||
| Multilingual Document Parsing | DocAtlas 1.0 (test) | 17 | May 14, 2026 | ||
| Multi-Armed Sampling | Multi-Armed Sampling Finite-Armed Setting theoretical bounds | 17 | May 14, 2026 | ||
| Long-text generation | LongText-Bench ZH | 17 | May 13, 2026 | ||
| Long-text generation | LongText-Bench EN | 17 | May 13, 2026 |