Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2007 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Remote Sensing Change Detection | LEVIRCD-MM | 17 | May 11, 2026 | ||
| Attribution | ImageNet-S (val) | 17 | May 8, 2026 | ||
| Speech enhancement | EARS-WHAM v2 (test) | 17 | May 8, 2026 | ||
| Offline Reinforcement Learning | OGBench p44 + ct aggregate | 17 | May 8, 2026 | ||
| Commonsense Reasoning | PIQA | 17 | May 8, 2026 | ||
| Question Answering | ARC Challenge | 17 | May 8, 2026 | ||
| Question Answering | OpenBookQA | 17 | May 8, 2026 | ||
| Commonsense Reasoning | Spoken StoryCloze | 17 | May 8, 2026 | ||
| Anomaly Detection | BTCV+LiTs BMAD benchmark (test) | 17 | May 8, 2026 | ||
| Edge Detection | PASCAL-Context | 17 | May 8, 2026 | ||
| Instruction | IFEval | 17 | May 19, 2026 | ||
| Locomotion | D4RL Cheetah Medium | 17 | May 19, 2026 | ||
| Emotion Recognition | SEED-V 5-Class | 17 | Jul 7, 2026 | ||
| UI-to-Code Generation | Design2Code | 17 | Jul 1, 2026 | ||
| Jailbreak Robustness | Mousetrap | 17 | May 7, 2026 | ||
| Jailbreak Robustness | AutoRAN | 17 | May 7, 2026 | ||
| Jailbreak Attack Robustness | GCG | 17 | May 7, 2026 | ||
| Multi-task Language Understanding | MMLU pass@1 | 17 | May 7, 2026 | ||
| Jailbreak Attack Defense | Mousetrap | 17 | May 7, 2026 | ||
| Jailbreak Attack Defense | AutoRAN | 17 | May 7, 2026 | ||
| Jailbreak Attack Defense | SorryBench | 17 | May 7, 2026 | ||
| Search | BrowseComp-ZH (test) | 17 | May 6, 2026 | ||
| Activation Steering | AxBench Gemma-2-9B layer 20 | 17 | May 6, 2026 | ||
| Multimodal Classification | NYU Depth V2 | 17 | Jun 1, 2026 | ||
| SVG Reconstruction | SVG | 17 | May 6, 2026 |