Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 909 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Spacecraft Pose Estimation | SPEED+ Lightbox | 29 | Jun 23, 2026 | ||
| Continuous Sign Language Recognition | Phoenix14 (dev) | 29 | Feb 26, 2026 | ||
| Neoadjuvant chemotherapy (NAC) response prediction | Multi-center TURBT slides (5-fold cross-val) | 29 | Feb 26, 2026 | ||
| Video Reasoning | SAGE-Bench 1.0 (test) | 29 | Feb 26, 2026 | ||
| Novel View Synthesis | DeepBlending | 29 | Jun 9, 2026 | ||
| Plant Disease Recognition | Citrus | 29 | Feb 26, 2026 | ||
| Episodic-memory Question Answering | OpenEQA v1 (ScanNet) | 29 | Feb 26, 2026 | ||
| Camera calibration and pose estimation | Collimator images | 29 | Feb 26, 2026 | ||
| 3D Reconstruction | NuScenes | 29 | Jul 8, 2026 | ||
| Visual Search | HR-Bench 8K | 29 | May 27, 2026 | ||
| Brain-structure segmentation | OASIS (test) | 29 | Feb 26, 2026 | ||
| Image Classification | APS Dataset | 29 | Feb 26, 2026 | ||
| Video Understanding and Reasoning | Video-MMMU (test) | 29 | Apr 23, 2026 | ||
| Pose Estimation | 7-Scenes | 29 | Apr 6, 2026 | ||
| Visual Grounding | RefFLIR 1.0 (val) | 29 | Feb 26, 2026 | ||
| text-to-image generation | PRISM | 29 | Jun 11, 2026 | ||
| Crack Segmentation | CRACK500 (test) | 29 | Jun 1, 2026 | ||
| Motion Deblur | ImageNet | 29 | May 19, 2026 | ||
| Color Constancy | NCC (in-dataset) | 29 | Feb 26, 2026 | ||
| Open-loop trajectory prediction | nuScenes v1.0 (test) | 29 | Feb 26, 2026 | ||
| Face Forgery Detection | DiFF Face Editing representative subset (test) | 29 | Apr 14, 2026 | ||
| Pose Estimation | Graz4K (test) | 29 | Feb 26, 2026 | ||
| Person Re-identification | AG-ReID G→A protocol v2 | 29 | Apr 30, 2026 | ||
| Person Re-identification | AG-ReID A→G protocol v2 | 29 | Apr 30, 2026 | ||
| Video Hallucination Evaluation | EventHallusion | 29 | Apr 15, 2026 |