Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 327 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Code correctness classification | LiveSQLBench SQLite | 55 | May 28, 2026 | ||
| AI-generated text detection | READ (test) | 55 | May 26, 2026 | ||
| Mathematical Reasoning | MATH | 55 | Jul 7, 2026 | ||
| Theory of Mind | ToMi | 55 | May 28, 2026 | ||
| Question Answering | Eight QA datasets Average | 55 | May 19, 2026 | ||
| Mathematical Reasoning | MathVista | 55 | May 20, 2026 | ||
| Information Retrieval | CLINC150 | 55 | May 13, 2026 | ||
| Information Retrieval | REALSCHOLAR | 55 | May 13, 2026 | ||
| Massive Multitask Language Understanding | MMLU | 55 | Jun 4, 2026 | ||
| Classification | Electricity (test) | 55 | Jun 1, 2026 | ||
| Adversarial Jailbreak Attack | Vicuna 13B | 55 | May 12, 2026 | ||
| Text-to-SQL | Spider Realistic | 55 | Jun 24, 2026 | ||
| ATE Estimation | Synth. data | 55 | May 7, 2026 | ||
| Jailbreak Attack Success Rate | HarmBench | 55 | Jun 16, 2026 | ||
| Watermark forensics | WIDERFace | 55 | Apr 30, 2026 | ||
| Image Reconstruction | CytoImageNet (test) | 55 | Apr 28, 2026 | ||
| Question Answering | NQ, TriviaQA, PopQA, HotpotQA, 2wiki, MuSiQue, Bamboogle | 55 | Jul 8, 2026 | ||
| Simulation-based Inference | Hierarchical Two Moons | 55 | Apr 23, 2026 | ||
| Simulation-based Inference | Hierarchical Gaussian Mixture | 55 | Apr 23, 2026 | ||
| Simulation-based Inference | Hierarchical Gaussian Linear Uniform | 55 | Apr 23, 2026 | ||
| Simulation-based Inference | Hierarchical Gaussian Linear | 55 | Apr 23, 2026 | ||
| Multi-Hop Question Answering | Bamboogle | 55 | May 20, 2026 | ||
| Event Prediction | RETWEET (test) | 55 | Jun 9, 2026 | ||
| Event Prediction | AMAZON (test) | 55 | Jun 9, 2026 | ||
| Event Prediction | STACKOVERFLOW (test) | 55 | Jun 9, 2026 |