SOTA2 Research · benchmarks
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
Benchmarks
309,131 indexed| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| High-Resolution Visual Perception | HRBench 8K | 29 | Jul 2, 2026 | ||
| Multi-hop Question Answering | MuSiQue (dev) | 29 | Jun 23, 2026 | ||
| Multimodal Information Seeking | InfoSeek | 29 | Jun 5, 2026 | ||
| Attribution Sensitivity Analysis | ImgNet 2012 (val) | 29 | Jun 5, 2026 | ||
| Fragment Retrieval | WeChat (test) | 29 | Jun 4, 2026 | ||
| Fragment Retrieval | MLDR (val) | 29 | Jun 4, 2026 | ||
| Room Navigation | Room Navigation Visual Appearance - VA | 29 | Jun 4, 2026 | ||
| Room Navigation | Room Navigation Complex Instruction - CI | 29 | Jun 4, 2026 | ||
| Room Navigation | Room Navigation Common Sense - CS | 29 | Jun 4, 2026 | ||
| Room Navigation | Room Navigation Base | 29 | Jun 4, 2026 | ||
| Node Classification | Cora 3-shot | 29 | Jun 11, 2026 | ||
| Document Parsing | OmniDocBench 1.6 (test) | 29 | Jun 4, 2026 | ||
| Mathematical reasoning | AIME 25 | 29 | Jun 30, 2026 | ||
| Mathematical reasoning | AIME 24 | 29 | Jun 30, 2026 | ||
| Novel View Synthesis | NVIDIA (test) | 29 | Jul 3, 2026 | ||
| Document Formatting | DOCFORMBENCH | 29 | Jun 2, 2026 | ||
| Harmful prompt detection | OpenAI | 29 | Jul 8, 2026 | ||
| Long-term forecasting | ETTm1 (test) | 29 | Jun 11, 2026 | ||
| Synthetic Network Generation | DBLP | 29 | Jun 2, 2026 | ||
| Synthetic Network Generation | YouTube | 29 | Jun 2, 2026 | ||
| Synthetic Network Generation | PolBlogs | 29 | Jun 2, 2026 | ||
| Image Classification | TissueMNIST MedMNIST v2 (test) | 29 | Jun 2, 2026 | ||
| Agentic Search | DeepSearchQA | 29 | Jun 23, 2026 | ||
| Text-to-Image | UniGenBench Long | 29 | Jun 2, 2026 | ||
| Text-to-Image | UniGenBench Short | 29 | Jun 2, 2026 |