Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 183 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Audio-to-text Retrieval | AudioCaps (test) | 80 | Jul 1, 2026 | ||
| Question Answering | TQA | 80 | Jun 1, 2026 | ||
| Logical Reasoning | ProofW | 80 | Feb 26, 2026 | ||
| Multilingual Information Retrieval | XQuAD | 80 | Jun 18, 2026 | ||
| Long-context language modeling | RULER | 80 | Jul 1, 2026 | ||
| Multi-turn Dialogue | MT-Bench | 80 | Apr 3, 2026 | ||
| Reasoning | LiveBench Reasoning | 80 | Feb 26, 2026 | ||
| Short-term time series forecasting | M4 Average | 80 | May 27, 2026 | ||
| Novel View Synthesis | Deep Blending | 80 | May 1, 2026 | ||
| Video Question Answering | MLVU | 80 | Jul 3, 2026 | ||
| Image-level Anomaly Detection | VisA | 80 | Apr 10, 2026 | ||
| Visual Speech Recognition | LRS3 High-Resource, 433h labelled v1 (test) | 80 | Feb 26, 2026 | ||
| 3D Object Detection | Place3D Adverse | 80 | Feb 26, 2026 | ||
| Visual Question Answering | SEED-Bench Image | 80 | Jun 9, 2026 | ||
| 3D Semantic Segmentation | ScanNet (S0) | 80 | Jul 1, 2026 | ||
| Few-shot 3D Scene Segmentation | ScanNet S1 | 80 | Mar 23, 2026 | ||
| Few-shot 3D Scene Segmentation | ScanNet S0 | 80 | Mar 23, 2026 | ||
| Semantic Segmentation | PASCAL-Context PC-459 | 80 | Jun 9, 2026 | ||
| Point Cloud Classification | ShapeNet Part | 80 | Feb 26, 2026 | ||
| Visual Question Answering | VQAv2 (test-dev) | 80 | Apr 6, 2026 | ||
| Interpretation Error Evaluation | ImageNet | 80 | Feb 26, 2026 | ||
| Mathematical Reasoning | MGSM (test) | 80 | Jun 2, 2026 | ||
| Continuous Sign Language Recognition | PHOENIX14-T (dev) | 80 | Mar 18, 2026 | ||
| OOD Detection | Textures (OOD) with CIFAR-10 (ID) (test) | 80 | Mar 4, 2026 | ||
| Author Profiling | PAN16 (test) | 80 | Feb 26, 2026 |