Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1001 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Brain Image Segmentation | MASiVar Cohort II | 27 | Apr 23, 2026 | ||
| Super-Resolution | DRealSR | 27 | Jun 2, 2026 | ||
| Commonsense Reasoning | Commonsense Reasoning | 27 | May 12, 2026 | ||
| Multimodal Question Answering | ManyModalQA (test) | 27 | Apr 23, 2026 | ||
| Adversarial Face Protection | Face protection evaluation set | 27 | Apr 21, 2026 | ||
| Mathematical Problem Solving | AMC | 27 | May 22, 2026 | ||
| Online Video Understanding | OVOBench 1.0 (test) | 27 | Apr 28, 2026 | ||
| Multi-task Language Understanding | MMLU | 27 | Jun 29, 2026 | ||
| Multimodal Benchmarking | MMBench EN | 27 | Jun 18, 2026 | ||
| Robotic Manipulation | SimplerEnv Google Robot Tasks Visual Matching and Variant Aggregation | 27 | Jun 2, 2026 | ||
| LLM Serving | High-load generation workload w/ NVLink | 27 | Apr 21, 2026 | ||
| LLM Serving | High-load generation workload w/o NVLink | 27 | Apr 21, 2026 | ||
| Anomaly Detection | UCR | 27 | Apr 21, 2026 | ||
| Image Classification | CIFAR10 Dirichlet-0.6 (test) | 27 | Apr 21, 2026 | ||
| Visual Question Answering (Closed-ended) | PathVQA closed-ended | 27 | Jun 8, 2026 | ||
| Long-term time series forecasting | ETTh2 Average | 27 | Jun 1, 2026 | ||
| Adversarial Attack Success Rate | Multi-agent Communication Scenarios (test) | 27 | Apr 21, 2026 | ||
| Molecular activity prediction | MoleculeACE (Aggregate) | 27 | Apr 20, 2026 | ||
| Dynamic Texture Recognition | DynTex++ | 27 | Apr 20, 2026 | ||
| Classification | LPD | 27 | Apr 20, 2026 | ||
| Time-Series Anomaly Detection | SWAT | 27 | Jun 23, 2026 | ||
| Robustness of Tool-use | RoTBench | 27 | Apr 16, 2026 | ||
| Tool-use Reasoning | Ours (test) | 27 | Apr 16, 2026 | ||
| Binary decision | TODAY | 27 | Apr 15, 2026 | ||
| Binary decision | PLASMA | 27 | Apr 15, 2026 |