Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 412 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Image Generation | MS-COCO 512x512 with Stable Diffusion (val) | 48 | Feb 26, 2026 | ||
| Image Generation | ImageNet 256x256 with FlowDCN (val) | 48 | Feb 26, 2026 | ||
| Image Generation | CIFAR-10 32x32 with ReFlow (test) | 48 | Feb 26, 2026 | ||
| Question Answering | ARC-Challenge 0-shot (test) | 48 | May 6, 2026 | ||
| Cooling load forecasting | DCData 100% train data 1.0 (test) | 48 | Feb 26, 2026 | ||
| Image Classification | CIFAR-10 (test) | 48 | Feb 26, 2026 | ||
| Image Completion | Standard Images SR 30% (test) | 48 | Feb 26, 2026 | ||
| Image Completion | Standard Images SR 20% (test) | 48 | Feb 26, 2026 | ||
| Image Classification | CIFAR100 | 48 | Feb 26, 2026 | ||
| Image Classification | CIFAR10 | 48 | Feb 26, 2026 | ||
| Hallucination Detection | GSM8K (test) | 48 | Feb 26, 2026 | ||
| Hallucination Detection | SQuAD (test) | 48 | Feb 26, 2026 | ||
| General Capability Evaluation | tinyBenchmarks | 48 | Feb 26, 2026 | ||
| Image Classification | ImageNet 1.0 (val) | 48 | Apr 10, 2026 | ||
| Role-Play Evaluation | CoSER | 48 | Jun 26, 2026 | ||
| Multi-answer Question Answering | MAQA-ΔK−1 | 48 | Feb 26, 2026 | ||
| Model Discovery | Llama-8B model tree Extended Discovery | 48 | Feb 26, 2026 | ||
| Model Discovery | Mistral-7B model tree Extended Discovery | 48 | Feb 26, 2026 | ||
| Model Discovery | Qwen-7B model tree (Extended Discovery) | 48 | Feb 26, 2026 | ||
| Model Discovery | Qwen-3B model tree Extended Discovery | 48 | Feb 26, 2026 | ||
| Contributor Attribution | Fashion Product | 48 | Feb 26, 2026 | ||
| Classification | National Health and Nutrition Examination Survey Employee 2013–2014 cycle (random sample splits) | 48 | Feb 26, 2026 | ||
| Classification | Diabetes | 48 | Feb 26, 2026 | ||
| Character-level Language Modeling | Shakespeare modern | 48 | Feb 26, 2026 | ||
| Distribution Learning | Gaussian distribution rho = 2 | 48 | Feb 26, 2026 |