Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2194 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Spatial Understanding | MMSI-Bench | 16 | Jun 11, 2026 | ||
| Valence Prediction | SEWA DB | 16 | Apr 10, 2026 | ||
| Arousal Prediction | SEWA DB | 16 | Apr 10, 2026 | ||
| Knowledge-Based Visual Question Answering | InfoSeek All | 16 | Apr 10, 2026 | ||
| Mathematical Reasoning | OlympicB | 16 | Apr 10, 2026 | ||
| Mathematical Reasoning | AIME 25 | 16 | Apr 10, 2026 | ||
| Mathematical Reasoning | AIME 24 | 16 | Apr 10, 2026 | ||
| ESG Sentiment Analysis | SloESG-News Aspect E 1.0 (test) | 16 | Apr 10, 2026 | ||
| Conversational Search | QReCC (test) | 16 | Apr 10, 2026 | ||
| Watermark Detection | Stable Diffusion Avg-Removal 2.1 | 16 | Apr 10, 2026 | ||
| Watermark Detection | Stable Diffusion Original 2-1 | 16 | Apr 10, 2026 | ||
| Energy and Force Prediction | rMD17 (test) | 16 | Apr 9, 2026 | ||
| Malicious Prompt Detection | JailbreakV_28K Text-based (test) | 16 | Apr 9, 2026 | ||
| Malicious Prompt Detection | JailbreakV_28K Image-based (test) | 16 | Apr 9, 2026 | ||
| Direct Malicious | VLSafe OOD | 16 | Apr 9, 2026 | ||
| Direct Malicious | MM-SafetyBench OOD | 16 | Apr 9, 2026 | ||
| Text-based Jailbreak | AdvBench-M OOD | 16 | Apr 9, 2026 | ||
| Text-based Jailbreak | JailbreakV_28K IND | 16 | Apr 9, 2026 | ||
| Image-based Jailbreak | HADES OOD | 16 | Apr 9, 2026 | ||
| Image-based Jailbreak | FigStep OOD | 16 | Apr 9, 2026 | ||
| Image-based Jailbreak | JailbreakV_28K IND | 16 | Apr 9, 2026 | ||
| Selective Prediction | 3 datasets (mean over all 21 runs) | 16 | Apr 9, 2026 | ||
| Multimodal Deep Search | FVQA | 16 | Apr 9, 2026 | ||
| Attack Success Rate Evaluation | 40 diverse smartphone tasks Total | 16 | Apr 9, 2026 | ||
| Attack Success Rate Evaluation | 40 diverse smartphone tasks Persuade subcategory | 16 | Apr 9, 2026 |