Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1353 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Knowledge Acquisition | TOFU author-profile questions (held-out) | 22 | May 20, 2026 | ||
| Multimodal Understanding | SEED-I, VizWiz, ScienceQA | 22 | May 20, 2026 | ||
| Text-to-Image Generation | DrawBench | 22 | May 20, 2026 | ||
| Gene mutation prediction | MUT-SETD2 MUT-HET-RCC | 22 | May 20, 2026 | ||
| Renal cell carcinoma staging | RCC-STAGING TCGA | 22 | May 20, 2026 | ||
| Breast cancer subtype classification | BRACS 3 | 22 | May 20, 2026 | ||
| Attribution Faithfulness | ImageNet (val) | 22 | May 20, 2026 | ||
| Few-Shot Semantic Segmentation | CD-FSS Average 5-shot | 22 | May 20, 2026 | ||
| Few-Shot Semantic Segmentation | Chest X-ray 5-shot | 22 | May 20, 2026 | ||
| Few-Shot Semantic Segmentation | Chest X-ray 1-shot | 22 | May 20, 2026 | ||
| Few-Shot Semantic Segmentation | ISIC 1-shot 2018 | 22 | May 20, 2026 | ||
| Few-Shot Semantic Segmentation | DeepGlobe 5-shot | 22 | May 20, 2026 | ||
| Few-Shot Semantic Segmentation | DeepGlobe 1-shot | 22 | May 20, 2026 | ||
| Mathematical Problem Solving | Apex Shortlist | 22 | Jul 2, 2026 | ||
| Image Classification | ImageNet-C severity level 5 v1 (test) | 22 | May 20, 2026 | ||
| Sequential Recommendation | MovieLens 1M (leave-one-out) | 22 | Jun 4, 2026 | ||
| Mathematical Reasoning | MATH 500 | 22 | Jul 2, 2026 | ||
| Zero-shot Reasoning | Downstream Tasks (LMB, PIQA, HellaSwag, OPQA, ARC) | 22 | May 20, 2026 | ||
| Evaluation | Core Metrics | 22 | May 20, 2026 | ||
| Question Answering | NQ | 22 | May 20, 2026 | ||
| Question Answering | TriviaQA | 22 | May 20, 2026 | ||
| Memory-Augmented GUI Interaction | MementoGUI-Bench | 22 | May 19, 2026 | ||
| Web Navigation | MM-Mind2Web | 22 | May 19, 2026 | ||
| GUI Navigation | GUI-Odyssey | 22 | May 28, 2026 | ||
| Articulated Object Modeling | AiP real | 22 | May 19, 2026 |