SOTA2 Research · benchmarks
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
Benchmarks
309,131 indexed| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Long-context Language Modeling | RULER 4K | 29 | Jun 1, 2026 | ||
| Optimization Modeling | NL4OPT | 29 | Jun 16, 2026 | ||
| Image Classification | 8-Vision benchmark (EuroSAT, GTSRB, SVHN, RESISC45, DTD, Cars, MNIST, SUN397) (test) | 29 | Jun 4, 2026 | ||
| Ambiguity Detection | AmbigQA | 29 | Jun 23, 2026 | ||
| Robotic Manipulation | LIBERO | 29 | Jul 9, 2026 | ||
| Salient Object Detection | DUT-OMRON 5168 images (test) | 29 | May 26, 2026 | ||
| Link Prediction | Products | 29 | May 26, 2026 | ||
| Code Generation | LiveCodeBench | 29 | Jun 18, 2026 | ||
| Mathematical Reasoning | MathVista | 29 | Jun 1, 2026 | ||
| Question Answering | NQ | 29 | May 25, 2026 | ||
| Iris Recognition | CASIA-Iris-Thousand | 29 | Jul 7, 2026 | ||
| Mathematical Reasoning | Mathematical Reasoning Suite MATH 500, AIME 2024, AIME 2025, AMC 2023, Olympiad Bench | 29 | Jun 2, 2026 | ||
| Binary Classification | binary health datasets avg (test) | 29 | May 19, 2026 | ||
| Shallow Water Equations Simulation | SWE | 29 | May 19, 2026 | ||
| 2D Diffusion-Reaction Simulation | DiffRe-2D | 29 | May 19, 2026 | ||
| Compressible Navier-Stokes Simulation | CompNS | 29 | May 19, 2026 | ||
| Compressible Euler Equations (Euler-2D) | VICON (test) | 29 | May 19, 2026 | ||
| Text Classification | NLU Tasks (SST-2, MR, Subj, SST-5, MPQA) | 29 | Jun 15, 2026 | ||
| Molecular Interaction Prediction | CombiSolv | 29 | May 19, 2026 | ||
| Irregular clinical time-series prediction | PhysioNet 2012 (test) | 29 | May 21, 2026 | ||
| General Knowledge Question Answering | MMLU-PRO | 29 | May 26, 2026 | ||
| Segmentation | LGG MRI | 29 | May 18, 2026 | ||
| Robot Manipulation | LIBERO Goal | 29 | May 20, 2026 | ||
| Modality Retrieval | Flickr30k | 29 | Jun 19, 2026 | ||
| Modality Retrieval | Toys | 29 | Jun 19, 2026 |