Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2278 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Language Identification | AfroScope High resource | 16 | Feb 26, 2026 | ||
| Legal Reasoning | LegalBench | 16 | Jun 1, 2026 | ||
| Logic Reasoning | Autologic en | 16 | Feb 26, 2026 | ||
| Logic Reasoning | Autologic cn | 16 | Feb 26, 2026 | ||
| Mathematical Reasoning | Minerva | 16 | Feb 26, 2026 | ||
| Mathematical Reasoning | MATH 500 | 16 | Feb 26, 2026 | ||
| STEM Reasoning | TheoremQA | 16 | Feb 26, 2026 | ||
| Science Reasoning | GPQA | 16 | Feb 26, 2026 | ||
| Response Classification | Generic Response Classification Suite (Aegis2.0, Beavertails, SEval, SafeRLHF, Think, WildG, XSTest) | 16 | Feb 26, 2026 | ||
| Prompt classification | XSTest | 16 | Feb 26, 2026 | ||
| Prompt classification | WildG | 16 | Feb 26, 2026 | ||
| Prompt classification | SorryB | 16 | Feb 26, 2026 | ||
| Prompt classification | SEval | 16 | Feb 26, 2026 | ||
| Prompt classification | OverR | 16 | Feb 26, 2026 | ||
| Prompt classification | OpenAIM | 16 | Feb 26, 2026 | ||
| Safety Evaluation | Attack Benchmarks Response | 16 | Feb 26, 2026 | ||
| Safety Evaluation | Attack Benchmarks Prompt | 16 | Feb 26, 2026 | ||
| Response Safety Classification | Multilingual Safety Benchmarks Response-side (test) | 16 | Feb 26, 2026 | ||
| Prompt Safety Classification | Multilingual Safety Benchmarks Prompt-side (test) | 16 | Feb 26, 2026 | ||
| Safe Completion | SEval Response 2.0 | 16 | Feb 26, 2026 | ||
| Safe Completion | SEval Prompt 2.0 | 16 | Feb 26, 2026 | ||
| Code Generation | MBPP | 16 | Feb 26, 2026 | ||
| Safety Alignment Evaluation | Refusal Evaluation Dataset | 16 | Feb 26, 2026 | ||
| Medical Image Classification | MedMNIST v2 (test) | 16 | Jun 8, 2026 | ||
| Discriminative Reasoning | RPEval Multi-MICRO Explicit Memory | 16 | Feb 26, 2026 |