Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1548 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Code Generation | BigCodeBench Lite-Pro Compositional Stream | 20 | Jun 2, 2026 | ||
| Angle-selection | MNIST-CT Large (test) | 20 | Jun 2, 2026 | ||
| Cross-modal Retrieval | CheXpert Plus | 20 | Jun 2, 2026 | ||
| Cross-modal Retrieval | MIMIC-CXR | 20 | Jun 2, 2026 | ||
| Safety Evaluation | MCV SafetyBench four-clip setting | 20 | Jun 2, 2026 | ||
| Multi-label Classification | FMoW (test) | 20 | Jun 2, 2026 | ||
| Map Alignment | ReplicaCAD Indoor Simulated | 20 | Jun 2, 2026 | ||
| Continual Learning | SuperNI (Order 2) | 20 | Jun 2, 2026 | ||
| Continual Learning | SuperNI (Order 1) | 20 | Jun 2, 2026 | ||
| Continual Learning | Long Sequence Order 2 | 20 | Jun 2, 2026 | ||
| Continual Learning | Long Sequence (Order 1) | 20 | Jun 2, 2026 | ||
| Situated Question Answering | SQA3D | 20 | Jul 8, 2026 | ||
| Software Engineering Task Success | SWE-Skills-Bench Hard | 20 | Jun 2, 2026 | ||
| Task Success | SkillLearnBench Random | 20 | Jun 2, 2026 | ||
| New Intent Discovery | CLINC | 20 | Jun 2, 2026 | ||
| Instructional Image Editing | CV-Arena | 20 | Jun 2, 2026 | ||
| Image Classification | TissueMNIST | 20 | Jun 2, 2026 | ||
| Functional Correctness | SMILES | 20 | Jun 2, 2026 | ||
| Functional Correctness | JSON | 20 | Jun 2, 2026 | ||
| Functional Correctness | C++ | 20 | Jun 2, 2026 | ||
| Multi-hop Question Answering | HotpotQA | 20 | Jun 2, 2026 | ||
| Video-grounded understanding and actionable reasoning | Pause-and-think-B 1.0 (test) | 20 | Jun 2, 2026 | ||
| Safety classification | Pre-Ex-Bench | 20 | Jun 2, 2026 | ||
| Safety classification | ASSEBench | 20 | Jun 2, 2026 | ||
| Multi-hop Question Answering | Multi-hop QA Suite (HotpotQA, 2Wiki, MuSiQue, G-Medical, G-Novel) | 20 | Jun 2, 2026 |