Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1715 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multimodal Affective Intelligence | AICA-Bench | 19 | Apr 8, 2026 | ||
| Therapeutic principle classification | FAITH-M | 19 | Apr 8, 2026 | ||
| Agentic Task Completion | τ2-Bench | 19 | Jun 4, 2026 | ||
| 3-year recurrence classification | Five TCGA cohorts UCEC LUAD LGG BRCA BLCA (average across cohorts) | 19 | Apr 8, 2026 | ||
| 3-year mortality classification | Five TCGA cohorts (UCEC, LUAD, LGG, BRCA, BLCA) (average across cohorts) | 19 | Apr 8, 2026 | ||
| Camera Pose Estimation | ETH3D | 19 | Apr 8, 2026 | ||
| Commonsense Question Answering | OBQA | 19 | Jun 2, 2026 | ||
| Tool-use agentic performance | AppWorld | 19 | Apr 7, 2026 | ||
| Tool-use agentic performance | BFCL V3 | 19 | Apr 7, 2026 | ||
| Multimodal Reasoning and Perception | MMStar (test) | 19 | Apr 7, 2026 | ||
| Image Captioning | Flickr30k | 19 | Apr 7, 2026 | ||
| Open World Video Understanding | VideoMMMU | 19 | Apr 7, 2026 | ||
| Temporal Spatial Reasoning | VSIBench | 19 | Apr 7, 2026 | ||
| Video Reasoning | STAR | 19 | Apr 14, 2026 | ||
| Question Answering | PopQA | 19 | Apr 7, 2026 | ||
| Numerical Question Answering | FinLongDocQA 1.0 (test) | 19 | Apr 7, 2026 | ||
| Backdoor Attack Detection | VPI | 19 | Apr 7, 2026 | ||
| Backdoor Attack Detection | BadNet | 19 | Apr 7, 2026 | ||
| Discourse-level Chinese-English translation | DiscoX | 19 | Apr 6, 2026 | ||
| Professional deep-research writing | Deepresearch-Gym | 19 | Apr 6, 2026 | ||
| Group Emotion Recognition | VGAF | 19 | Jun 9, 2026 | ||
| Pose Estimation | ScanNet v2 | 19 | May 21, 2026 | ||
| Crystal structure prediction | MPTS-52 | 19 | Apr 28, 2026 | ||
| Crystal structure prediction | MP-20 | 19 | Apr 28, 2026 | ||
| Mathematical Reasoning | MATH-500 | 19 | May 11, 2026 |