Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 876 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Discriminative Performance | HH_Golden 42.5k (test) | 30 | Feb 26, 2026 | ||
| Generative Performance | Ultrafeedback 61.1k (test) | 30 | Feb 26, 2026 | ||
| Discriminative Performance | Ultrafeedback 61.1k (test) | 30 | Feb 26, 2026 | ||
| Reliability Assessment | Strategic probe sets | 30 | Feb 26, 2026 | ||
| Influence Detection | Fashionpedia | 30 | Feb 26, 2026 | ||
| Safety Evaluation | PS-Bench base setting (test) | 30 | Feb 26, 2026 | ||
| GUI planning | AndroidControl High | 30 | May 19, 2026 | ||
| Function Calling | ToolBench Average | 30 | May 29, 2026 | ||
| Mathematical Reasoning | MATH 500 | 30 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K | 30 | Feb 26, 2026 | ||
| Online Bin Packing Problem | BPP online N=5k, W=100 | 30 | Mar 23, 2026 | ||
| Time Series Forecasting | ILI (test) | 30 | Mar 26, 2026 | ||
| Ranking | Candidate Profiles | 30 | Feb 26, 2026 | ||
| Reasoning | AIME 24 | 30 | Apr 28, 2026 | ||
| Brain network classification | ABCD-OCD | 30 | Feb 26, 2026 | ||
| Hallucination Detection | EventHallusion | 30 | Feb 26, 2026 | ||
| General Question Answering | NQ (Natural Questions) in-domain (val/test) | 30 | Jun 9, 2026 | ||
| Training Throughput Measurement | LIBERO | 30 | Feb 26, 2026 | ||
| Deep Research | xbench | 30 | Feb 26, 2026 | ||
| Clinical Question Answering | MedQA | 30 | Jul 3, 2026 | ||
| Scientific Olympiad Reasoning | FrontierScience-Olympiad | 30 | May 14, 2026 | ||
| Counterfactual Reasoning | MMLU-CF | 30 | Feb 26, 2026 | ||
| Univariate Forecasting | ETTh1 v1 (test) | 30 | Feb 26, 2026 | ||
| 3D Super-Resolution | NeRF Synthetic | 30 | Jul 3, 2026 | ||
| Tamper Localization | NIST | 30 | Jun 24, 2026 |