Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1636 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Cross-Corpus Ranking | Cross-Corpus Dataset | 20 | Feb 26, 2026 | ||
| General Language Understanding | 12-task evaluation suite (test) | 20 | Feb 26, 2026 | ||
| Long-form text generation | FactScore | 20 | Jul 9, 2026 | ||
| Question Answering | NQ | 20 | Feb 26, 2026 | ||
| Confidence Estimation | TRICKME | 20 | Feb 26, 2026 | ||
| Confidence Estimation | GRACE | 20 | Feb 26, 2026 | ||
| Confidence Estimation | GUESS | 20 | Feb 26, 2026 | ||
| Confidence Estimation | 20Q | 20 | Feb 26, 2026 | ||
| Multimodal Autoformalization | MATHVERSE Solid Geometry | 20 | Feb 26, 2026 | ||
| Sentiment Classification | SST2 | 20 | Feb 27, 2026 | ||
| Sentiment Classification | IMDb | 20 | Feb 27, 2026 | ||
| Machine Unlearning | TOFU 5% forget | 20 | Feb 26, 2026 | ||
| DB Task | Lifelong Agent Bench DB Task | 20 | Jun 17, 2026 | ||
| Safety Evaluation | TrustLLM | 20 | Mar 18, 2026 | ||
| Question Answering | HotpotQA 296:204 (test) | 20 | Feb 26, 2026 | ||
| Question Answering | QASPER 1200:251 (test) | 20 | Feb 26, 2026 | ||
| Preference Alignment | PRISM | 20 | Feb 26, 2026 | ||
| General Knowledge | MMLU-Pro | 20 | Apr 28, 2026 | ||
| Prompt Steering | Prompt Steering | 20 | Feb 26, 2026 | ||
| Efficiency-aware code generation | EffiBench-X C++ | 20 | Feb 26, 2026 | ||
| Efficiency-aware code generation | EffiBench-X Python | 20 | Feb 26, 2026 | ||
| Knowledge & Reasoning | TriviaQA | 20 | Apr 21, 2026 | ||
| Multimodal Conversation | PCogAlignBench (LS2) | 20 | Feb 26, 2026 | ||
| Multimodal Conversation | PCogAlignBench LS1 | 20 | Feb 26, 2026 | ||
| Multimodal Conversation | MMRole OOD | 20 | Feb 26, 2026 |