Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 949 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Text-to-Image Generation | R2I-Bench | 28 | Mar 31, 2026 | ||
| Scientific Reasoning | GPQA | 28 | Mar 20, 2026 | ||
| PDE Solving | PDEBench rdb (test) | 28 | Mar 20, 2026 | ||
| Rollout Prediction | PDEBench rdb (test) | 28 | Mar 20, 2026 | ||
| Rollout Prediction | PDEBench diff-sorp (test) | 28 | Mar 20, 2026 | ||
| Visual Perception | OCRBench | 28 | Jun 16, 2026 | ||
| Building Extraction | WHU dataset | 28 | Mar 20, 2026 | ||
| Video Understanding | Aggregate Video Benchmarks Suite | 28 | Apr 8, 2026 | ||
| Video Compression | UVG | 28 | Jun 23, 2026 | ||
| Video Super-Resolution | RealVSR | 28 | May 28, 2026 | ||
| 3D Object Detection | K-Radar v1 (test) | 28 | Mar 18, 2026 | ||
| Step Verification | VisualProcessBench (test) | 28 | May 18, 2026 | ||
| Light Field Super-Resolution | LFSR Average | 28 | Mar 18, 2026 | ||
| Omnidirectional Visual Understanding | 360Bench 1.0 (test) | 28 | Mar 18, 2026 | ||
| Public opinion simulation | WVS (World Values Survey) (test) | 28 | Mar 18, 2026 | ||
| Public Opinion Simulation | WVS | 28 | Mar 18, 2026 | ||
| Commonsense Reasoning | TruthfulQA | 28 | Apr 8, 2026 | ||
| Open-Vocabulary Instance Segmentation | MARIS in-domain (val) | 28 | Mar 18, 2026 | ||
| LLM Inference Performance | Synthetic Poisson Process Requests | 28 | Mar 18, 2026 | ||
| Audio Question Answering | MMAU | 28 | Apr 28, 2026 | ||
| Visible-Infrared Person Re-identification | LLCM Infrared2Visible | 28 | Jun 11, 2026 | ||
| Time Series Anomaly Detection | PUMP | 28 | Apr 16, 2026 | ||
| Multimodal Understanding | MME | 28 | Mar 18, 2026 | ||
| Face Manipulation Detection | FF++ High Quality (test) | 28 | Mar 17, 2026 | ||
| Video-level A/H Recognition | BAH (test) | 28 | Mar 19, 2026 |