Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1739 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-Discipline Reasoning | MMMU | 19 | Feb 26, 2026 | ||
| Table-to-Image Generation | TableVisBench v1 (test) | 19 | Feb 26, 2026 | ||
| Knowledge-based Visual Question Answering | INFOSEEK (Unseen Entity) | 19 | Apr 10, 2026 | ||
| Autonomous Driving | Bench2Drive base (train) | 19 | Feb 26, 2026 | ||
| Binary classification | MMAD | 19 | Feb 26, 2026 | ||
| Overall Survival Prediction | TCGA-LUSC Lung Squamous Cell Carcinoma N = 434 (5-fold cross-validation) | 19 | Feb 26, 2026 | ||
| Mathematical Reasoning | MathVista Math | 19 | Feb 26, 2026 | ||
| Embodied Question Answering | Episodic Memory EQA (EM-EQA) | 19 | Apr 7, 2026 | ||
| Video Recommendation | MicroLens 100k (test) | 19 | Feb 26, 2026 | ||
| Novel View Synthesis | ACID 16-view (test) | 19 | Mar 27, 2026 | ||
| Few-shot Object Counting | FSC147 1.0 (test) | 19 | Feb 26, 2026 | ||
| Few-shot Object Counting | FSC147 1.0 (val) | 19 | Feb 26, 2026 | ||
| Image Anomaly Detection | MVTec AD Hole | 19 | Feb 26, 2026 | ||
| Ship Re-identification | HOSS-ReID (All) | 19 | Feb 26, 2026 | ||
| Anomaly Detection and Localization | MU-Glioma-Post (test) | 19 | Feb 26, 2026 | ||
| Image Classification | Camelyon 17 (test) | 19 | Apr 27, 2026 | ||
| Salient Object Detection | VT1000 | 19 | Feb 26, 2026 | ||
| Chart Understanding | ChartQA | 19 | Feb 26, 2026 | ||
| Image Classification | CelebA WILDS (test) | 19 | Mar 20, 2026 | ||
| Text-to-Image Generation | MS-COCO 10K prompts 2014 (val) | 19 | Feb 26, 2026 | ||
| Lifelong Person Re-identification | CUHK03 | 19 | Feb 26, 2026 | ||
| Lifelong Person Re-identification | MSMT17 | 19 | Feb 26, 2026 | ||
| Lifelong Person Re-identification | DukeMTMC | 19 | Feb 26, 2026 | ||
| Lifelong Person Re-identification | CUHK-SYSU | 19 | Feb 26, 2026 | ||
| Lifelong Person Re-identification | Market1501 | 19 | Feb 26, 2026 |