Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 741 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Audio Retrieval | Clotho | 33 | May 19, 2026 | ||
| Image Classification | ImageNet 1K (val) | 33 | Feb 26, 2026 | ||
| Scene Text Recognition | ICDAR 2013 | 33 | Jun 2, 2026 | ||
| Scene Text Recognition | SVT 647 images | 33 | Feb 26, 2026 | ||
| Knowledge Base Question Answering | GrailQA v1.0 (test) | 33 | Feb 26, 2026 | ||
| Object Detection | Clipart (test) | 33 | Mar 24, 2026 | ||
| Code Generation | MBPP v1 (test) | 33 | May 13, 2026 | ||
| Text-to-Video Retrieval | SS label v2 | 33 | Apr 7, 2026 | ||
| Few-shot Image Classification | StanfordCars | 33 | Jun 5, 2026 | ||
| 5-way Few-shot Classification | tiered-ImageNet (test) | 33 | Feb 26, 2026 | ||
| Trajectory Prediction | NFL Football Dataset (NFL) 2017 | 33 | Feb 26, 2026 | ||
| Image Classification | CIFAR-10 | 33 | Mar 10, 2026 | ||
| Image Reconstruction | MS-COCO 2017 (val) | 33 | Jun 1, 2026 | ||
| Link Prediction | PCQM-Contact LRGB (test) | 33 | Feb 26, 2026 | ||
| Node Classification | COCO-SP LRGB (test) | 33 | Apr 7, 2026 | ||
| Grasp Detection | GraspNet-1Billion RealSense Novel | 33 | Mar 18, 2026 | ||
| Grasp Detection | GraspNet-1Billion RealSense Similar | 33 | Mar 18, 2026 | ||
| Grasp Detection | GraspNet-1Billion RealSense (Seen) | 33 | Mar 18, 2026 | ||
| Text Recognition | Chinese text recognition benchmark | 33 | Feb 26, 2026 | ||
| Instance Segmentation | MS COCO 2014 (val) | 33 | Feb 26, 2026 | ||
| Offline Reinforcement Learning | D4RL walker2d full-replay v2 | 33 | Jun 2, 2026 | ||
| Offline Reinforcement Learning | D4RL hopper-full-replay v2 | 33 | Jun 2, 2026 | ||
| Optimal Auction Design | Multi-setting Auction mixture of 2x3 to 3x7 settings (test) | 33 | Feb 26, 2026 | ||
| Toxicity Evaluation | RealToxicityPrompts | 33 | Jun 4, 2026 | ||
| Numerical Question Answering | FinQA (test) | 33 | Feb 27, 2026 |