Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1083 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Blur Prediction | CIFAR-10 | 26 | Feb 26, 2026 | ||
| Crop Prediction | CIFAR-10 | 26 | Feb 26, 2026 | ||
| Class-conditional Image Generation | ImageNet 256 | 26 | Jul 8, 2026 | ||
| Medical Image Classification | Mura s | 26 | Feb 26, 2026 | ||
| Medical Image Classification | AD | 26 | Feb 26, 2026 | ||
| Science Question Answering | ScienceQA v1.0 (test) | 26 | Feb 26, 2026 | ||
| Question Answering | HotpotQA v1.1 (test) | 26 | Feb 26, 2026 | ||
| Machine Translation | WMT ZH-EN 22 | 26 | Apr 30, 2026 | ||
| Image Classification | CIFAR-10 512-image subset (test) | 26 | Feb 27, 2026 | ||
| Panoramic Image Quality Assessment | OIQA In-dataset | 26 | Mar 17, 2026 | ||
| Analogy Generation | E-KAR | 26 | Feb 26, 2026 | ||
| Anomaly Detection | MVTec LOCO AD Structural Anomalies | 26 | Feb 26, 2026 | ||
| Few-Shot Classification | CIFAR-FS (test) | 26 | May 15, 2026 | ||
| Visual Localization | Cambridge Landmarks Hospital | 26 | Apr 8, 2026 | ||
| Image Captioning | COCO (test) | 26 | Apr 2, 2026 | ||
| Trajectory Prediction | ETH/UCY ETH, Hotel, Univ, Zara1, Zara2 | 26 | Feb 26, 2026 | ||
| Depth Estimation | DDAD | 26 | Jul 8, 2026 | ||
| Semantic Segmentation | GTAV + Synthia (G+S) to BDD100K (B), Cityscapes (C), Mapillary (M) Sim-2-Real (test) | 26 | Mar 31, 2026 | ||
| Caption Matching and Retrieval | NoCaps (val) | 26 | Feb 26, 2026 | ||
| Domain Adaptation | DomainNet-126 | 26 | Feb 26, 2026 | ||
| Semantic Segmentation | Cityscapes original (1/8 (372 labeled images)) | 26 | Jul 7, 2026 | ||
| Image Classification | SVHN (test) | 26 | Feb 26, 2026 | ||
| Deblurring | Deblurring (test) | 26 | Jul 3, 2026 | ||
| Multi-rater Medical Image Segmentation | LIDC-IDRI (test) | 26 | May 12, 2026 | ||
| Continual Learning | CIFAR100 10-task sequential (test) | 26 | Mar 4, 2026 |