Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2492 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Text Generation | TextGen | 15 | Feb 26, 2026 | ||
| Natural Language Processing | NLP | 15 | Feb 26, 2026 | ||
| Image Classification | ImgCls | 15 | Feb 26, 2026 | ||
| Mathematical reasoning | AIME24 | 15 | Apr 15, 2026 | ||
| Temporal Link Prediction | Googlemap CT inductive | 15 | May 8, 2026 | ||
| Temporal Link Prediction | Googlemap CT transductive | 15 | May 8, 2026 | ||
| Continuous Control | Hopper v5 | 15 | Apr 13, 2026 | ||
| Sampling on discretised synthetic densities | ManyWell (d=32) | 15 | Jun 1, 2026 | ||
| Code Generation | MBPP base standard (test) | 15 | Feb 26, 2026 | ||
| Mathematical Problem Solving | MATH standard (val) | 15 | Feb 26, 2026 | ||
| Mathematical Reasoning | Math-MC (test) | 15 | Feb 26, 2026 | ||
| Commonsense Reasoning | HellaSwag published (test) | 15 | Feb 26, 2026 | ||
| Image Classification | MNIST (test) | 15 | Feb 26, 2026 | ||
| Language Modeling | Chinchilla 2022 (train) | 15 | Feb 26, 2026 | ||
| Clustering | BACH | 15 | Feb 26, 2026 | ||
| Clustering | CHEN | 15 | Feb 26, 2026 | ||
| Protein Backbone Generation | SCOPe | 15 | Feb 26, 2026 | ||
| Conditional Independence Testing | Post-nonlinear CI benchmark synthetic n=1000 (dz=50) | 15 | Feb 26, 2026 | ||
| Conditional Independence Testing | Post-nonlinear synthetic CI benchmark (dz=10, n=1000) | 15 | Feb 26, 2026 | ||
| Classification | FRANKENSTEIN F3→F2 | 15 | Feb 26, 2026 | ||
| Classification | FRANKENSTEIN F3→F1 | 15 | Feb 26, 2026 | ||
| Classification | FRANKENSTEIN F3→F0 | 15 | Feb 26, 2026 | ||
| Classification | FRANKENSTEIN F2→F3 | 15 | Feb 26, 2026 | ||
| Classification | FRANKENSTEIN F2→F1 | 15 | Feb 26, 2026 | ||
| Classification | FRANKENSTEIN F2→F0 | 15 | Feb 26, 2026 |