Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 650 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Camera pose estimation | SUN3D Unknown Scene | 36 | Feb 26, 2026 | ||
| Camera pose estimation | YFCC100M (Unknown Scene) | 36 | Feb 26, 2026 | ||
| Mathematical Reasoning | MathVista MVistam | 36 | May 5, 2026 | ||
| Text-to-Image Generation | TIIF | 36 | May 12, 2026 | ||
| Single-object 4D Motion Generation | User Study Single-object 4D Motion Generation 1.0 (test) | 36 | Feb 26, 2026 | ||
| 3D Medical Image Synthesis | 3D MRI (test) | 36 | Feb 26, 2026 | ||
| Medical Image Segmentation | BraTS 2019 (10% labeled data) | 36 | Jul 3, 2026 | ||
| Multimodal Classification | FOOD101 UPMC (train test) | 36 | Feb 26, 2026 | ||
| Multimodal Classification | CUB (train test) | 36 | Feb 26, 2026 | ||
| Multimodal Classification | ROSMAP (train test) | 36 | Feb 26, 2026 | ||
| Multimodal Classification | BRCA (train test) | 36 | Feb 26, 2026 | ||
| Few-shot classification | PatternNet (19-19) | 36 | Feb 26, 2026 | ||
| Input selection | ImageNet (unlabeled pool) | 36 | Feb 26, 2026 | ||
| Accuracy Improvement | Cifar-10 (val) | 36 | Feb 26, 2026 | ||
| Classical Image Super-Resolution | Manga109 classical SR (test) | 36 | Feb 26, 2026 | ||
| Out-of-Distribution Detection | ImageNet-1K (ID) vs Textures (OOD) (test) | 36 | Jul 8, 2026 | ||
| Image Generation | Sub-IN 256x256 (test) | 36 | Feb 26, 2026 | ||
| Video Classification | Drive&Act | 36 | Feb 26, 2026 | ||
| Frame-level Deepfake Detection | DFDC | 36 | Apr 21, 2026 | ||
| Classification | MIL brain MRI dataset (test) | 36 | Feb 26, 2026 | ||
| Classification | fastMRI prostate | 36 | Feb 26, 2026 | ||
| Classification | fastMRI knee | 36 | Feb 26, 2026 | ||
| Object Detection | Foggy Cityscapes F (test) | 36 | Feb 26, 2026 | ||
| Absolute Trajectory Estimation | TUM RGB-D | 36 | Apr 27, 2026 | ||
| Depth Super-Resolution / Completion | KITTI (test) | 36 | Feb 26, 2026 |