Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2121 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Visual Instruction Following | RefCOCO | 16 | Jun 19, 2026 | ||
| Visual Instruction Following | SUN Instructive Visual Benchmark | 16 | Jun 19, 2026 | ||
| Visual Instruction Following | Caltech 101 Instructive Visual Benchmark | 16 | Jun 19, 2026 | ||
| Visual Instruction Following | ImageNet Instructive Visual Benchmark | 16 | Jun 19, 2026 | ||
| Top-1 Retrieval | RefCOCO Gemini Oracle (test) | 16 | Jun 19, 2026 | ||
| Top-1 Retrieval | SUN Gemini Annotations Oracle (test) | 16 | Jun 19, 2026 | ||
| Top-1 Retrieval | Caltech 101 Gemini Oracle (test) | 16 | Jun 19, 2026 | ||
| Top-1 Retrieval | ImageNet Gemini Oracle (test) | 16 | Jun 19, 2026 | ||
| Classification | CIFAR-10 | 16 | Jun 23, 2026 | ||
| Time-series forecasting | Time-MMD | 16 | Jun 19, 2026 | ||
| Protein Fold Classification | SCOP Fold (test) | 16 | Jun 19, 2026 | ||
| Continuous Control | LunarLanderContinuous v3 | 16 | Jun 19, 2026 | ||
| Subjective Search Question Answering | Internal Evaluation Subjective Q&A | 16 | Jun 19, 2026 | ||
| Semantic Segmentation | PANGAEA Evaluation protocol | 16 | Jun 19, 2026 | ||
| Video Understanding | VideoMME w/o sub. Long | 16 | Jun 18, 2026 | ||
| Text-to-Image Preference Evaluation | Aesthetic | 16 | Jun 18, 2026 | ||
| Text-to-Image Preference Evaluation | HPS v2 | 16 | Jun 18, 2026 | ||
| Text-to-Image Preference Evaluation | ImageReward | 16 | Jun 18, 2026 | ||
| Mathematics | MVision pt-PT | 16 | Jun 18, 2026 | ||
| Captioning | RefCap pt-PT | 16 | Jun 18, 2026 | ||
| Captioning | COCO pt-PT | 16 | Jun 18, 2026 | ||
| Spatial Understanding | RefRec pt-PT | 16 | Jun 18, 2026 | ||
| Spatial Understanding | EmbSp pt-PT | 16 | Jun 18, 2026 | ||
| Chart & Diagram Understanding | ChartQA pt-PT | 16 | Jun 18, 2026 | ||
| Chart & Diagram Understanding | AI2D pt-PT | 16 | Jun 18, 2026 |