Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 387 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Semantic Segmentation | ADE20K (test) | 50 | Feb 26, 2026 | ||
| Permuted Sequential Image Classification | MNIST Permuted Sequential | 50 | Feb 26, 2026 | ||
| Traveling Salesperson Problem | TSP-100 | 50 | Jun 19, 2026 | ||
| Retrosynthesis | USPTO-50k Reaction type known (test) | 50 | Feb 26, 2026 | ||
| Semantic Segmentation | DDD17 | 50 | Mar 5, 2026 | ||
| Class-conditional image generation | ImageNet-1K 256x256 (test) | 50 | Feb 26, 2026 | ||
| Text-to-SQL | Spider-Realistic | 50 | Jun 5, 2026 | ||
| Instruction Following | SelfInst | 50 | Feb 26, 2026 | ||
| Video Anomaly Detection | ShanghaiTech standard (test) | 50 | Feb 26, 2026 | ||
| Human Motion Prediction | Human3.6M | 50 | Jun 8, 2026 | ||
| Continual Learning | Large Number of Tasks | 50 | Feb 26, 2026 | ||
| Text Classification | PubMed | 50 | Feb 26, 2026 | ||
| Document Summarization | GovReport (test) | 50 | Feb 26, 2026 | ||
| Anomaly Detection | BraTS 2021 | 50 | Mar 4, 2026 | ||
| Future edge prediction | Enron (inductive) | 49 | Jul 7, 2026 | ||
| transductive future edge prediction | MOOC | 49 | Jul 7, 2026 | ||
| Classification | Analysis N-shot | 49 | Jun 30, 2026 | ||
| Mathematical Reasoning | MATH 500 | 49 | Jun 26, 2026 | ||
| Efficiency Benchmarking | Llama 8B base variant on H200 3.1 | 49 | Jun 23, 2026 | ||
| Constraint Following | ConstraintRot (test) | 49 | Jun 23, 2026 | ||
| Long context retrieval and reasoning | RULER | 49 | Jul 1, 2026 | ||
| Point Cloud Part Segmentation | ModelNet40 | 49 | Jun 11, 2026 | ||
| Point Cloud Segmentation | SemanticKITTI v2 (test) | 49 | Jun 11, 2026 | ||
| Two-Sample Testing | CIFAR10-WRN8 | 49 | Jun 9, 2026 | ||
| Two-Sample Testing | BLOB | 49 | Jun 9, 2026 |