Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1704 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Mathematical Reasoning | Math Benchmarks AIME 2024 AIME 2025 OlympiadBench | 19 | May 13, 2026 | ||
| Code Generation | Code Benchmarks LCBv6 & MBPP+ | 19 | May 13, 2026 | ||
| Cross-view Geo-localization | DenseUAV Drone → Satellite | 19 | May 15, 2026 | ||
| Backlit Image Enhancement | Backlit300 | 19 | May 13, 2026 | ||
| Helpfulness Prediction | Amazon-MRHP Home | 19 | May 13, 2026 | ||
| Helpfulness Prediction | Amazon-MRHP Electronics | 19 | May 13, 2026 | ||
| Helpfulness Prediction | Amazon-MRHP Clothing | 19 | May 13, 2026 | ||
| Helpfulness Prediction | Lazada-MRHP Home | 19 | May 13, 2026 | ||
| Helpfulness Prediction | Lazada-MRHP Electronics | 19 | May 13, 2026 | ||
| Helpfulness Prediction | Lazada-MRHP Clothing | 19 | May 13, 2026 | ||
| Forward-direction molecular property editing | MolEditRL 500 molecules (test) | 19 | May 12, 2026 | ||
| Sentence Classification | Combined dataset (K1, K2, K3, R4) | 19 | May 12, 2026 | ||
| Node Classification | Texas 1-shot | 19 | May 12, 2026 | ||
| Node Classification | Computers 1-shot | 19 | May 12, 2026 | ||
| Node Classification | Cornell 1-shot | 19 | May 12, 2026 | ||
| Node Classification | Wisconsin 1-shot | 19 | May 12, 2026 | ||
| Visual Grounding | Visual Grounding benchmarks (HRBench8K, VStarBench, RWQA, MMStar, GQA, MMBench v1.1, BLINK) | 19 | May 12, 2026 | ||
| Forward PDE problem solving | Burgers’ | 19 | May 12, 2026 | ||
| Class-Incremental Learning | AZ-Class IL | 19 | May 12, 2026 | ||
| Class-Incremental Learning | EMBER Class-IL | 19 | May 12, 2026 | ||
| Satisfiability Solving | Leaking Bucket | 19 | May 12, 2026 | ||
| Simple P-log Program Computation | dice | 19 | May 12, 2026 | ||
| Video Summarization | TVSum (5FCV) | 19 | May 12, 2026 | ||
| Science Question Answering | SQA | 19 | May 12, 2026 | ||
| Visual Question Answering | VizWiz | 19 | May 12, 2026 |