Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1123 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Ad detection | Ad detection dataset Old LLMs 1.0 | 25 | Mar 6, 2026 | ||
| Targeted Adversarial Attack | Flickr30K 1,000 images (test) | 25 | Mar 6, 2026 | ||
| Targeted Adversarial Attack | Flickr30K | 25 | Mar 6, 2026 | ||
| Link Prediction | Polblogs | 25 | May 28, 2026 | ||
| Multimodal Understanding | MMMU (dev) | 25 | May 4, 2026 | ||
| Out-of-Distribution Detection | UCF101 OOD | 25 | Apr 14, 2026 | ||
| Language Modeling | LAMBADA | 25 | Apr 7, 2026 | ||
| Question Answering | OfficeQA | 25 | Mar 4, 2026 | ||
| Tokenization | WikiText-103 | 25 | Mar 4, 2026 | ||
| Graph Classification | DBLP v1 | 25 | Mar 4, 2026 | ||
| Gaussian mean estimation | Gaussian Model n=1000 | 25 | Mar 4, 2026 | ||
| Multi-turn Conversation Evaluation | MT-bench | 25 | Mar 4, 2026 | ||
| Crack Segmentation | DeepCrack (test) | 25 | Jun 1, 2026 | ||
| Composed Image Retrieval | Fashion-IQ Toptee (val) | 25 | Mar 4, 2026 | ||
| Composed Image Retrieval | Fashion-IQ Dress (val) | 25 | Mar 4, 2026 | ||
| Question Answering | G-bench Novel | 25 | Jun 2, 2026 | ||
| Motion deblur | FFHQ 256x256 | 25 | Jun 30, 2026 | ||
| Jailbreak Attack | MM-SafetyBench (tiny) | 25 | Mar 4, 2026 | ||
| Long-form writing | LongBench-Write (test) | 25 | Jun 4, 2026 | ||
| Visual Question Answering | SQA-Image | 25 | Feb 27, 2026 | ||
| Multimodal Classification | CREMA-D (test) | 25 | Apr 8, 2026 | ||
| Hyperspectral Anomaly Detection | Hyperion | 25 | Jun 17, 2026 | ||
| Image Classification | CIFAR100 | 25 | Apr 7, 2026 | ||
| Classification | 16 datasets average | 25 | Mar 25, 2026 | ||
| Class-incremental learning | ImageNet100 | 25 | May 28, 2026 |