Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 979 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Neural Architecture Search | HW-NAS-Bench Edge GPU | 28 | Feb 26, 2026 | ||
| Landmark Regression | MAFL (test) | 28 | Feb 26, 2026 | ||
| Out-of-Distribution Detection | CIFAR-10 in-dist ImageNet out-dist | 28 | Feb 26, 2026 | ||
| Few-shot Image Classification | MiniImageNet 5-way 1-shot | 28 | Feb 26, 2026 | ||
| Text Classification | GLUE (val) | 28 | Jul 3, 2026 | ||
| Object Detection | COCO standard 5% labeled 2017 (train) | 28 | Feb 26, 2026 | ||
| Collaborative Human Motion Prediction | ExPI (single action split) | 28 | Feb 26, 2026 | ||
| Image Inpainting | CelebA-HQ 256x256 (test) | 28 | May 8, 2026 | ||
| Space-Time Video Super-Resolution | Adobe-Center (test) | 28 | Apr 21, 2026 | ||
| Space-Time Video Super-Resolution | GoPro Center (test) | 28 | Apr 21, 2026 | ||
| Length-of-Stay prediction | MIMIC-III | 28 | Jun 5, 2026 | ||
| Anomaly Detection | UCR | 28 | Jun 1, 2026 | ||
| Text-to-Video Retrieval | ActivityNet (val1) | 28 | Feb 26, 2026 | ||
| Classification | German Credit (test) | 28 | Apr 10, 2026 | ||
| Image Classification | ImageNet-1K 1.0 (1% labels) | 28 | Feb 26, 2026 | ||
| Novel View Synthesis | LLFF real-world scenes 17 (test) | 28 | Feb 26, 2026 | ||
| Affordance Grounding | AGD20k v1 (Unseen) | 28 | Jul 7, 2026 | ||
| Affordance Grounding | AGD20k v1 (Seen) | 28 | Jul 7, 2026 | ||
| Face Anti-Spoofing | CASIA-MFSD O&M&I to C (test) | 28 | Feb 26, 2026 | ||
| Image Classification | CIFAR-100 | 28 | Jun 30, 2026 | ||
| Visual Object Tracking | NFS 30 FPS | 28 | Feb 26, 2026 | ||
| Transition Video Question Answering | TGIF-QA (test) | 28 | Feb 26, 2026 | ||
| Co-Salient Object Detection | CoCA (test) | 28 | Feb 26, 2026 | ||
| Motion Deblurring | RealBlur-R raw (test) | 28 | Feb 26, 2026 | ||
| Few-Shot Classification | Cars (test) | 28 | Feb 26, 2026 |