Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 2184 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Video Generation | User Study | 16 | May 29, 2026 | ||
| Attack Detection | ATRDF2 | 16 | Apr 23, 2026 | ||
| Commonsense Reasoning | PIQA | 16 | Apr 23, 2026 | ||
| Question Answering | ARC Challenge | 16 | Apr 23, 2026 | ||
| Deep Information Retrieval and Research | xbench DeepSearch | 16 | Apr 23, 2026 | ||
| Web Browsing and Navigation (Chinese) | BrowseComp-ZH | 16 | Apr 23, 2026 | ||
| Multi-view Crowd Tracking | Wildtrack | 16 | Apr 23, 2026 | ||
| Multitask Evaluation | Aggregate (MMLU, GSM8K, SQuAD V2, Amazon, 20News) | 16 | Apr 23, 2026 | ||
| Sentiment Analysis | Amazon | 16 | Apr 23, 2026 | ||
| Face Anti-Spoofing | MICO Cross-dataset | 16 | Apr 23, 2026 | ||
| Multiple-choice QA | Multiple-choice QA Benchmarks (PIQA, OpenBookQA, HellaSwag, ARC) | 16 | Apr 23, 2026 | ||
| Image Classification | CIFAR-100 standard (test) | 16 | May 28, 2026 | ||
| TSG Execution | 80 real-world incidents | 16 | Apr 23, 2026 | ||
| Optimal Power Flow prediction | PGLearn full topology | 16 | Apr 23, 2026 | ||
| Multi-temporal Semantic Segmentation (Crop Type) | Crop Type Mapping | 16 | Apr 23, 2026 | ||
| Scientific Reasoning | GPQA (test) | 16 | Apr 23, 2026 | ||
| Length of Stay | eICU | 16 | Apr 23, 2026 | ||
| Mathematical Problem Solving | MATH | 16 | May 19, 2026 | ||
| Face Reconstruction | AgeDB | 16 | Apr 21, 2026 | ||
| Facial Template Inversion | LFW (test) | 16 | Apr 21, 2026 | ||
| Recommendation | MovieLens | 16 | May 26, 2026 | ||
| Time Series Forecasting | Rock four time intervals ahead | 16 | Apr 21, 2026 | ||
| Time Series Forecasting | NYCT four time intervals ahead | 16 | Apr 21, 2026 | ||
| Time Series Forecasting | NYCT one time interval ahead | 16 | Apr 21, 2026 | ||
| Allocative Efficiency Prediction | Double Auction AE Round 2+ (test) | 16 | Apr 21, 2026 |