Loading the SOTA2 catalog…
SOTA2 Research · benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
| Task | Dataset | Domain | Trend | Results | Last Update |
|---|---|---|---|---|---|
| Math Reasoning | AIME 2024 | 61 | Jun 26, 2026 | ||
| Document Retrieval | MuSiQue (test) | 61 | Jun 30, 2026 | ||
| Role-play dialogue comprehension | SocialBench | 61 | May 26, 2026 | ||
| Video Reasoning | LongVideoReason | 61 | Jun 24, 2026 | ||
| Node classification | CS | 61 | Jun 2, 2026 | ||
| Question Answering |
| Bamboogle |
| 61 |
| May 29, 2026 |
| Humanoid Object Search | H*Bench | 61 | Jul 8, 2026 |
|---|
| Mathematical Reasoning | MinervaMath | 61 | May 29, 2026 |
|---|
| Class-conditional Image Generation | ImageNet class-conditional 256x256 | 61 | May 26, 2026 |
|---|
| Harmful Content Detection | UnsafeBench | 61 | Jun 25, 2026 |
|---|
| Hidden value recovery | American Housing Survey (AHS) Minneapolis–St. Paul–Bloomington | 61 | Apr 27, 2026 |
|---|
| Sequential Recommendation | Sports | 61 | May 28, 2026 |
|---|
| Multiple-choice tasks | FrenchBench | 61 | Mar 16, 2026 |
|---|
| Mathematical Reasoning | GSM8K | 61 | May 29, 2026 |
|---|
| Math Reasoning | GSM8K | 61 | Jun 16, 2026 |
|---|
| Commonsense Reasoning | Winogrande | 61 | Jun 26, 2026 |
|---|
| Commonsense Reasoning | ARC | 61 | Jun 1, 2026 |
|---|
| Question Answering | SpotterQA | 61 | Mar 9, 2026 |
|---|
| Multiple-choice Question Answering | OpenBookQA (test) | 61 | May 28, 2026 |
|---|
| Camera Pose Estimation | CO3D v2 (test) | 61 | Jul 8, 2026 |
|---|
| Re-identification | RGBNT201 | 61 | Jun 19, 2026 |
|---|
| Writing | WritingBench v1 (test) | 61 | May 1, 2026 |
|---|
| Question Answering | qasper | 61 | Feb 27, 2026 |
|---|
| Constrained Incremental Graph Drawing | Newly generated instances | 61 | Feb 26, 2026 |
|---|
| Medical Image Classification | PathMNIST | 61 | Jun 23, 2026 |
|---|