Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 872 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multimodal Reasoning | O3-BENCH (test) | 30 | Feb 26, 2026 | ||
| Localization | RefCOCO (testB) | 30 | Apr 6, 2026 | ||
| Localization | RefCOCO (testA) | 30 | Apr 6, 2026 | ||
| Localization | RefCOCO (val) | 30 | Apr 6, 2026 | ||
| Scene Text Understanding | OCRBench | 30 | Jul 7, 2026 | ||
| Classification | Forest River South Bound (train) | 30 | Feb 26, 2026 | ||
| Classification | Forest River South Bound (test) | 30 | Feb 26, 2026 | ||
| Classification | Park River North Bound (train) | 30 | Feb 26, 2026 | ||
| Classification | Park River North Bound (test) | 30 | Feb 26, 2026 | ||
| Multimodal Evaluation | MME | 30 | Jul 1, 2026 | ||
| Image Classification | ImageNet-1K | 30 | Feb 26, 2026 | ||
| Phase retrieval | FFHQ | 30 | May 28, 2026 | ||
| Relative Pose Estimation | SUN3D (test) | 30 | Feb 26, 2026 | ||
| Relative Pose Estimation | PhotoTourism (test) | 30 | Feb 26, 2026 | ||
| Reasoning Image Editing | RISEBench 1.0 (test) | 30 | Feb 26, 2026 | ||
| Object Hallucination Evaluation | MSCOCO (val) | 30 | Jun 1, 2026 | ||
| Gaussian Deblur | FFHQ | 30 | May 21, 2026 | ||
| Hallucination Detection | COCO-Caption (val) | 30 | Feb 26, 2026 | ||
| Hallucination Detection | M-HalDetect (val) | 30 | Feb 26, 2026 | ||
| Hallucination Detection | AMBER sampled 5k | 30 | Feb 26, 2026 | ||
| Accuracy Improvement | ImageNet (val) | 30 | Feb 26, 2026 | ||
| Image Retrieval | ImageNet seen classes @1000 | 30 | Feb 26, 2026 | ||
| Image Retrieval | CIFAR-10 @1000 (seen classes) | 30 | Feb 26, 2026 | ||
| Image Retrieval | ImageNet @1000 15%-unseen | 30 | Feb 26, 2026 | ||
| Image Retrieval | CIFAR-10 (@1000) 20%-unseen | 30 | Feb 26, 2026 |