Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1948 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Multi-Robot Task Allocation | LEGO-MRTA (test) | 18 | Feb 26, 2026 | ||
| Context-aware Instruction Following | Context-aware Instructions (test) | 18 | Feb 26, 2026 | ||
| Text-based Person Retrieval | UFine3C (evaluation) | 18 | Feb 26, 2026 | ||
| Instruction Following | AlpacaEval GPT-4 (test) | 18 | Apr 29, 2026 | ||
| Video Representation | UVG (test) | 18 | Feb 26, 2026 | ||
| Video representation | UVG | 18 | Feb 26, 2026 | ||
| Instruction Following | SuperNI Hold-In v1.0 (test) | 18 | Feb 26, 2026 | ||
| Instruction Following | SuperNI Hold-Out v1.0 (test) | 18 | Feb 26, 2026 | ||
| Question Answering | NarrativeQA No Trun. latest (test) | 18 | Feb 26, 2026 | ||
| Multi-modal Benchmark | MMBench Chinese (dev) | 18 | Feb 26, 2026 | ||
| Scene Classification | EuroSAT Sentinel-2 (train val) | 18 | Feb 26, 2026 | ||
| Question Answering | HotpotQA (test) | 18 | Feb 26, 2026 | ||
| Table-to-text Generation | Logic2Text (test) | 18 | Feb 26, 2026 | ||
| Image Classification | Small-ImageNet-127 size 32x32 and 64x64 (test) | 18 | Feb 26, 2026 | ||
| Reinforcement Learning | MuJoCo Humanoid v2 | 18 | Feb 26, 2026 | ||
| Reinforcement Learning | MuJoCo Ant v2 | 18 | Feb 26, 2026 | ||
| Reinforcement Learning | MuJoCo Walker2d v2 | 18 | Feb 26, 2026 | ||
| Reinforcement Learning | MuJoCo Hopper v2 | 18 | Feb 26, 2026 | ||
| Reinforcement Learning | MuJoCo HalfCheetah v2 | 18 | Feb 26, 2026 | ||
| Point Cloud Filtering | PCNet 10K Sparse Laplace noise (synthetic) | 18 | Feb 26, 2026 | ||
| Point Cloud Filtering | PCNet Sparse 10K | 18 | Feb 26, 2026 | ||
| Point Cloud Filtering | PUNet Dense 50K | 18 | Feb 26, 2026 | ||
| Semantic Segmentation | EndoVis 17 | 18 | Feb 26, 2026 | ||
| Interactive Segmentation | COD10K | 18 | Apr 15, 2026 | ||
| Region-Level Captioning | refCOCOg (test) | 18 | Feb 26, 2026 |