Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1903 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Fine-grained species classification | Weeds Species196 16-shot (test) | 18 | Feb 26, 2026 | ||
| Fine-grained species classification | Insecta Species196 16-shot (test) | 18 | Feb 26, 2026 | ||
| Fine-grained species classification | iNaturalist Aves 16-shot 2018 (test) | 18 | Feb 26, 2026 | ||
| Knowledge-Intensive Reasoning | 2wikiMultiHopQA | 18 | Feb 27, 2026 | ||
| Knowledge-Intensive Reasoning | WebWalker | 18 | Feb 27, 2026 | ||
| Clustering | Phishing 5K | 18 | Feb 26, 2026 | ||
| Text-to-Image Generation | MS-COCO 512x512 with SDv3.5 | 18 | Feb 26, 2026 | ||
| Image Generation | ImageNet 256x256 with FlowDCN | 18 | Feb 26, 2026 | ||
| Image Generation | CIFAR-10 32x32 with ReFlow | 18 | Feb 26, 2026 | ||
| Electrogram reconstruction from Body Surface Potential Maps | BSPM-EGM (test) | 18 | Feb 26, 2026 | ||
| Policy Optimization | Office World MAP0 | 18 | Feb 26, 2026 | ||
| Image Classification | CIFAR-10 | 18 | Feb 26, 2026 | ||
| Function Calling | TB-MM | 18 | Feb 26, 2026 | ||
| Function Calling | TB-HF | 18 | Feb 26, 2026 | ||
| Function Calling | TB-DL | 18 | Feb 26, 2026 | ||
| Function Calling | TinyAgent | 18 | Feb 26, 2026 | ||
| Tool Retrieval and Function Selection | Taskbench-HF | 18 | Feb 26, 2026 | ||
| Role-Playing | RPGBench Aggregate (Overall) | 18 | Feb 26, 2026 | ||
| Role-Playing | RPGBench Dialogue Shift (Generalization) | 18 | Feb 26, 2026 | ||
| Role-Playing | RPGBench Character Shift (Generalization) | 18 | Feb 26, 2026 | ||
| Role-Playing | RPGBench User Shift Generalization | 18 | Feb 26, 2026 | ||
| Role-Playing | RPGBench In-distribution | 18 | Feb 26, 2026 | ||
| Traffic Prediction | PeMS-BAY 15 min horizon (test) | 18 | Feb 26, 2026 | ||
| Image Classification | ImageNet V2 | 18 | Feb 26, 2026 | ||
| EEG Classification | BCIC IV-2A (test) | 18 | Apr 10, 2026 |