Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1112 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Jailbreaking | AdvBench 20% evaluation | 25 | Apr 23, 2026 | ||
| Harmfulness Detection | Aegis | 25 | Jun 2, 2026 | ||
| Action Recognition | NTU RGB+D X-View | 25 | Jul 2, 2026 | ||
| Conversation Summarization | Bitcoin | 25 | Apr 21, 2026 | ||
| Conversation Summarization | Stack | 25 | Apr 21, 2026 | ||
| Conversation Summarization | 25 | Apr 21, 2026 | |||
| Text-to-Image Generation | UNIGENBENCH (test) | 25 | Apr 21, 2026 | ||
| Anomaly Detection | TSB-AD 180 series | 25 | Apr 21, 2026 | ||
| Scientific reasoning | SciQ | 25 | May 8, 2026 | ||
| Scientific Reasoning | GPQA | 25 | Apr 21, 2026 | ||
| Mathematical Reasoning | AIME25 | 25 | Apr 21, 2026 | ||
| Document Text Transcription | Malvern-Hills-5+ | 25 | Apr 21, 2026 | ||
| Video classification | UCF-101 (val) | 25 | Jun 25, 2026 | ||
| General Visual Question Answering | RealWorldQA | 25 | Jul 7, 2026 | ||
| Image Captioning | Image Captioning | 25 | May 1, 2026 | ||
| Hallucination Detection | HaluEval (in-distribution) | 25 | Jun 30, 2026 | ||
| Cognitive and Psychological tasks | Parent data (25-fold cross-val) | 25 | Apr 20, 2026 | ||
| Document-level tasks | Parent data (25-fold cross-val) | 25 | Apr 20, 2026 | ||
| Cognitive and Psychological spectrum prediction | offspring data 25 fold (val) | 25 | Apr 20, 2026 | ||
| Document-level traits prediction | offspring data 25 fold cross-validation | 25 | Apr 20, 2026 | ||
| Claim-level rating prediction | Insq | 25 | Apr 20, 2026 | ||
| Claim-level rating prediction | FORA | 25 | Apr 20, 2026 | ||
| Ultrasound Image Segmentation | TN3K (test) | 25 | Jul 3, 2026 | ||
| Scalability analysis | Digits | 25 | Apr 16, 2026 | ||
| Preference Prediction | RewardBench | 25 | Jun 4, 2026 |