Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1641 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Question Answering | NQ | 20 | Feb 26, 2026 | ||
| Automatic Speech Recognition | LSVSC (test) | 20 | May 28, 2026 | ||
| Scientific Table & Figure Analysis | ANABENCH 1.0 (test) | 20 | Feb 26, 2026 | ||
| Scientific Table & Figure Analysis | ANABENCH C.4 (test) | 20 | Feb 26, 2026 | ||
| World Modeling | ScienceWorld | 20 | Feb 26, 2026 | ||
| World Modeling | Plancraft | 20 | Feb 26, 2026 | ||
| World Modeling | Webshop (test) | 20 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K (test) | 20 | Feb 26, 2026 | ||
| Multi-Hop Question Answering | Multi-Hop QA Average | 20 | Feb 26, 2026 | ||
| Single-Hop Question Answering | NQ (Natural Questions) in-domain (test) | 20 | Feb 26, 2026 | ||
| Code | LCB v6 | 20 | Jun 2, 2026 | ||
| Fact-checking | FEVEROUS (test) | 20 | Feb 26, 2026 | ||
| Safety Evaluation | PKU-SafeRLHF-V | 20 | Feb 26, 2026 | ||
| Multimodal Deception Detection | MM-DeceptionBench | 20 | Feb 26, 2026 | ||
| Graph Coloring | Random Graph Coloring | 20 | Feb 26, 2026 | ||
| Traveling Salesman Problem | TSP N=20 (test) | 20 | Mar 27, 2026 | ||
| LoRA Retrieval | CARLoS LoRA Retrieval Evaluation Set (test) | 20 | Feb 26, 2026 | ||
| Scaffold decoration | CrossDock-SBDD | 20 | Feb 26, 2026 | ||
| Scaffold decoration | CrossDock Weak | 20 | Feb 26, 2026 | ||
| Crash-severity assessment | NEISS Subset | 20 | Feb 26, 2026 | ||
| Deep Research | BrowseComp-EN (BC-en) original (test) | 20 | Feb 26, 2026 | ||
| Deep Research | GAIA text-only original (test) | 20 | Feb 26, 2026 | ||
| Trajectory-level safety evaluation | ATBench (test) | 20 | Feb 26, 2026 | ||
| Trajectory-level safety evaluation | ASSE-Safety (test) | 20 | Feb 26, 2026 | ||
| GUI Task Execution | AITZ | 20 | Mar 26, 2026 |