Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 910 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Object Hallucination | A-OKVQA POPE (test) | 29 | Mar 9, 2026 | ||
| Optic Disc segmentation | Refuge | 29 | Apr 24, 2026 | ||
| Optic Cup Segmentation | Refuge | 29 | Apr 24, 2026 | ||
| Medical Image Segmentation | ACDC (5% labeled) | 29 | Feb 26, 2026 | ||
| Visual Question Answering | GQA | 29 | Mar 17, 2026 | ||
| Action Recognition | DailyDVS-200 | 29 | May 21, 2026 | ||
| Chart-based Reasoning | CharXivRQ | 29 | Apr 10, 2026 | ||
| Low-light Video Enhancement | BVI-RLV (test) | 29 | Apr 28, 2026 | ||
| Industrial Anomaly Detection | MMAD one-shot 1.0 (test) | 29 | Feb 26, 2026 | ||
| Human Activity Recognition | Skoda | 29 | May 20, 2026 | ||
| Robotic Manipulation (Goal) | LIBERO-Goal 1.0 (test) | 29 | Jun 23, 2026 | ||
| Robotic Manipulation (Object) | LIBERO-Object 1.0 (test) | 29 | Jun 23, 2026 | ||
| Robotic Manipulation (Spatial) | LIBERO-Spatial 1.0 (test) | 29 | Jun 23, 2026 | ||
| Secret information extraction | Cat | 29 | Feb 26, 2026 | ||
| Secret information extraction | Bedroom | 29 | Feb 26, 2026 | ||
| Secret information extraction | FFHQ | 29 | Feb 26, 2026 | ||
| Misinformation Detection | Chinese Dataset | 29 | May 13, 2026 | ||
| Misinformation Detection | English Dataset | 29 | May 13, 2026 | ||
| General Evaluation | AGIEval | 29 | Mar 17, 2026 | ||
| Multi-task Reasoning | Average (2WikiMultiHop, MMLU, GSM8k) (in-distribution) | 29 | Apr 21, 2026 | ||
| Safety Classification | Wildguardmix | 29 | Jun 23, 2026 | ||
| Safety Risk Detection | internal Agentic AI workflow benchmark | 29 | Feb 26, 2026 | ||
| Long-context Retrieval | Needle-in-a-Haystack | 29 | Jun 2, 2026 | ||
| Mathematical Reasoning | OlympBench | 29 | May 21, 2026 | ||
| Medical Knowledge Performance | Polish Board Certification Examinations (test) | 29 | Feb 26, 2026 |