Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1906 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Correctness Prediction | MMLU | 18 | Feb 26, 2026 | ||
| Correctness Prediction | TruthQA | 18 | Feb 26, 2026 | ||
| Correctness Prediction | MedQA | 18 | Feb 26, 2026 | ||
| Correctness Prediction | LogiQA | 18 | Feb 26, 2026 | ||
| Correctness Prediction | MathQA | 18 | Feb 26, 2026 | ||
| Correctness Prediction | Overall Combined Datasets | 18 | Feb 26, 2026 | ||
| Math Reasoning | MATH | 18 | Apr 28, 2026 | ||
| Mathematical Reasoning | Aggregate MATH, Minerva, Olympiad, GAO, AMC23, AIME24, AIME25 | 18 | Feb 26, 2026 | ||
| Mathematical Reasoning | AIME 25 | 18 | Feb 27, 2026 | ||
| Mathematical Reasoning | GAO | 18 | Feb 26, 2026 | ||
| Electricity Forecasting | UK | 18 | Feb 26, 2026 | ||
| Synthetic Regression | Syn-10 | 18 | Feb 26, 2026 | ||
| Time Series Forecasting | TimeMMD Energy | 18 | Mar 25, 2026 | ||
| Continuous Control | Hopper 3-Dof | 18 | Feb 26, 2026 | ||
| Node Classification | Amazon-Computer non-overlapping (5 clients) | 18 | May 8, 2026 | ||
| Image Classification | ImageNet-1K (test) | 18 | Feb 26, 2026 | ||
| Machine Unlearning | TOFU Llama-3.2 Instruct (average of 1%, 5%, 10% forget sets) | 18 | Feb 26, 2026 | ||
| Machine Unlearning | TOFU 10% forget set 1.0 | 18 | Feb 26, 2026 | ||
| Machine Unlearning | TOFU 1% (forget set) | 18 | Feb 26, 2026 | ||
| Contributor Attribution | ArtBench Post-Impressionism (test) | 18 | Feb 26, 2026 | ||
| Mathematical Reasoning | Math MATH500, AIME24, Minerva-Math, AMC23 | 18 | Feb 26, 2026 | ||
| Scientific Reasoning | Science Domain In-Domain: SampleQA, GPQA(ALL), HLE | 18 | Feb 26, 2026 | ||
| Classification | Circles Manifold 3D (test) | 18 | Feb 26, 2026 | ||
| Function Fitting | Holography Dataset (test) | 18 | Feb 26, 2026 | ||
| OOD Detection | WMT15 En-Fr nt2014 OOD ID (test) | 18 | Feb 26, 2026 |