Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2006 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Single-Doc Question Answering | LongBench-E Single-Doc QA | 17 | May 12, 2026 | ||
| Radar-Camera Depth Estimation | nuScenes 80m range v1.0-trainval (val) | 17 | May 12, 2026 | ||
| Classification | blood-transfusion | 17 | May 12, 2026 | ||
| Automated Parking | Standard Slot | 17 | May 26, 2026 | ||
| Embodied Household Task | ALFWorld (test) | 17 | May 12, 2026 | ||
| Scene Description | WAD (test) | 17 | May 12, 2026 | ||
| 3D Question Answering | SQA3D (val) | 17 | Jul 3, 2026 | ||
| Abductive Logical Reasoning | QUAIL | 17 | May 11, 2026 | ||
| Unconditional Generation | FineWeb-Edu Mask source 170M-parameter (train) | 17 | May 11, 2026 | ||
| Unconditional Generation | FineWeb-Edu Uniform source 170M-parameter (train) | 17 | May 11, 2026 | ||
| Video Question Answering | OVO-Bench | 17 | May 11, 2026 | ||
| Video Question Answering | OVO-Bench Backward Tracing | 17 | May 11, 2026 | ||
| Video Question Answering | OVO-Bench Real-Time Visual Perception | 17 | May 11, 2026 | ||
| Mathematical Reasoning | LG-GSM8K | 17 | May 11, 2026 | ||
| Deepfake Detection | DaGAN FFHQ | 17 | May 11, 2026 | ||
| Deepfake Detection | DiffSwap FFHQ | 17 | May 11, 2026 | ||
| Deepfake Detection | DeeperForensics | 17 | May 11, 2026 | ||
| Mathematical Reasoning | HMMT25 Nov. | 17 | May 11, 2026 | ||
| PDE Solving | Darcy Flow 256x256 resolution | 17 | May 11, 2026 | ||
| PDE Solving | Darcy Flow 128x128 resolution | 17 | May 11, 2026 | ||
| PDE Solving | Darcy Flow 32x32 resolution | 17 | May 11, 2026 | ||
| Node Classification | Cora expanded (test) | 17 | May 11, 2026 | ||
| Causal Discovery | DREAM3 | 17 | Jun 12, 2026 | ||
| Remote Sensing Change Detection | Gaza-Change MM v2 (test) | 17 | May 11, 2026 | ||
| Remote Sensing Change Detection | WHUCD-MM | 17 | May 11, 2026 |