Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1107 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Text-to-Image Retrieval | SARVLM-1M (test) | 25 | May 18, 2026 | ||
| Image-to-Text Retrieval | SARVLM-1M (test) | 25 | May 18, 2026 | ||
| Molecular prediction | BACE full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | HIV full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | MUV full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | ClinTox full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | SIDER full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | ToxCast full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | Tox21 full-shot (test) | 25 | May 18, 2026 | ||
| Molecular prediction | BBBP full-shot (test) | 25 | May 18, 2026 | ||
| Protein function prediction | PPI full-shot (test) | 25 | May 18, 2026 | ||
| Molecular property prediction | PPI 50-shot (test) | 25 | May 18, 2026 | ||
| Reasoning | Average (MATH500, AMC23, AIME24, AIME25, GPQA-d) | 25 | May 15, 2026 | ||
| Science Reasoning | GPQA diamond | 25 | May 15, 2026 | ||
| Mathematical Reasoning | AIME 25 | 25 | May 15, 2026 | ||
| Prompt Injection and Jailbreak Detection | EIA | 25 | May 15, 2026 | ||
| Rental price regression | Melbourne Airbnb (test) | 25 | May 15, 2026 | ||
| Reinforcement Learning | MuJoCo Hopper target planet (Venus-like gravity) | 25 | May 15, 2026 | ||
| Unlearning | TOFU (Split99) | 25 | May 15, 2026 | ||
| Temporal Video Grounding | Charades-STA | 25 | Jun 30, 2026 | ||
| Math | MATH 500 | 25 | Jun 2, 2026 | ||
| Runtime Safety Monitoring | Crossroad | 25 | May 15, 2026 | ||
| Scene Graph Generation | PSG v1.0 (test) | 25 | May 14, 2026 | ||
| Visual Question Answering | GQA 40 (test) | 25 | May 14, 2026 | ||
| EEG Classification | TUEV | 25 | Jun 23, 2026 |