Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 161 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Code Generating | MBPP | 88 | Feb 26, 2026 | ||
| Safety Alignment | HarmBench | 88 | Feb 26, 2026 | ||
| Object Detection | BDD100K | 88 | Apr 30, 2026 | ||
| Safety Evaluation | DirectHarm 4 | 87 | Jun 2, 2026 | ||
| Zero-shot Classification | Downstream Tasks Zero-shot (BoolQ, HellaSwag, WinoGrande, ARC-e, ARC-c, PIQA, OBQA) | 87 | Apr 16, 2026 | ||
| Lipschitz Constant Estimation | Small MLP networks (test) | 87 | Apr 9, 2026 | ||
| Mathematics | AIME 2024 | 87 | Jun 30, 2026 | ||
| Image Classification | DTD | 87 | Jun 4, 2026 | ||
| Embodied AI Task Planning | EB-ALFRED | 87 | Jun 11, 2026 | ||
| Code Generation | MBPP | 87 | Jun 26, 2026 | ||
| Category Discovery | CUB-200 2011 | 87 | May 1, 2026 | ||
| Mathematical Reasoning | Minerva Math v1 (test) | 87 | Jun 2, 2026 | ||
| Mathematical Visual Question Answering | MathVista | 87 | May 12, 2026 | ||
| Audio Deepfake Detection | ASVspoof DF 2021 | 87 | Jul 1, 2026 | ||
| Time Series Forecasting | ETTm1 | 87 | Mar 13, 2026 | ||
| Visible-Infrared Person Re-Identification | RegDB Thermal2Visible v1 | 87 | Feb 26, 2026 | ||
| Image Classification | UCM (test) | 87 | Mar 9, 2026 | ||
| Classification | DTD | 87 | Jul 7, 2026 | ||
| Class-incremental learning | CIFAR-100 10-split | 87 | Jun 23, 2026 | ||
| GUI Grounding | MMBench-GUI L2 (test) | 87 | Jul 1, 2026 | ||
| Commonsense Reasoning | Commonsense Reasoning Suite (test) | 87 | Jun 9, 2026 | ||
| Text-to-Speech | Seed-TTS zh (test) | 87 | Jun 9, 2026 | ||
| Graph Anomaly Detection | Photo | 87 | Jun 30, 2026 | ||
| Conditional Image Generation | ImageNet 64x64 (val) | 87 | Mar 25, 2026 | ||
| Open Vocabulary Semantic Segmentation | COCO Object with background | 87 | May 13, 2026 |