Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 827 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Misalignment Detection | Workbench n=62 (general) | 30 | Jul 3, 2026 | ||
| Misalignment Detection | Agent-SafetyBench underspecified queries n=148 | 30 | Jul 3, 2026 | ||
| Node Classification | Products-CL | 30 | Jul 2, 2026 | ||
| Text Generation | OpenWebText | 30 | Jul 2, 2026 | ||
| Text-to-Image Generation | LongTextBench | 30 | Jul 1, 2026 | ||
| Text-to-Image Generation | TextScenesHQ | 30 | Jul 1, 2026 | ||
| Super-Resolution | DIV2K 2017 (test) | 30 | Jun 30, 2026 | ||
| Tabular Classification | Anemia | 30 | Jun 30, 2026 | ||
| Parameter estimation | Synthetic AR(5) stationary process 1M sample size (test) | 30 | Jun 29, 2026 | ||
| Artifact Removal | Synthetic Artifacts Cityscapes-based | 30 | Jun 29, 2026 | ||
| Web Navigation | WebVoyager OOD | 30 | Jun 26, 2026 | ||
| Safety Classification | AEGIS 2 | 30 | Jul 8, 2026 | ||
| Multimodal Performance Evaluation | MME perception | 30 | Jun 30, 2026 | ||
| Classification | ImageNet 30 20-10 class split Closed-set (test) | 30 | Jun 26, 2026 | ||
| Fact Verification | FEVER | 30 | Jun 25, 2026 | ||
| Knowledge Conflict Resolution | Conflicts | 30 | Jun 25, 2026 | ||
| MM 2D Grounding | CountBenchQA | 30 | Jul 7, 2026 | ||
| Code Understanding | CRUX (test) | 30 | Jun 24, 2026 | ||
| Formal Logic Reasoning | MMLU formal_logic (test) | 30 | Jun 24, 2026 | ||
| Text generation | 1024 samples | 30 | Jun 24, 2026 | ||
| Text generation | uniform DFM backbone 1024 samples | 30 | Jun 24, 2026 | ||
| Backdoor Attack | SIDER | 30 | Jun 23, 2026 | ||
| Backdoor Attack | BBBP | 30 | Jun 23, 2026 | ||
| Air Traffic Flow and Capacity Management | ATFCM Scenarios (aggregated) | 30 | Jun 23, 2026 | ||
| Classification | 3D Normal Data d=3 | 30 | Jun 23, 2026 |