Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1171 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Simulator selection | COVID-19 | 24 | Jun 4, 2026 | ||
| Policy Selection | Measles | 24 | Jun 4, 2026 | ||
| Policy Selection | Supply Chain | 24 | Jun 4, 2026 | ||
| Policy Selection | COVID-19 | 24 | Jun 4, 2026 | ||
| Mathematical Reasoning | AIME 2024 | 24 | Jul 9, 2026 | ||
| Class-conditional Image Generation | CIFAR-10 class-conditional (test) | 24 | Jun 4, 2026 | ||
| Video Instance Segmentation | YouTubeVIS 2019 | 24 | Jun 15, 2026 | ||
| Logical Reasoning | LogiQA2 | 24 | Jun 4, 2026 | ||
| JSON Fuzzing | JSON | 24 | Jun 4, 2026 | ||
| Fuzzing | JSON fuzzing target | 24 | Jun 4, 2026 | ||
| Machine Learning Engineering | MLE-Dojo | 24 | Jun 4, 2026 | ||
| Scientific Discovery | ScienceAgentBench | 24 | Jun 4, 2026 | ||
| Data Analysis Code Generation | DA-Code | 24 | Jun 4, 2026 | ||
| Common-sense Reasoning | lm-eval-harness 11 tasks | 24 | Jun 4, 2026 | ||
| Language Modeling | Nemotron-CC 25M held-out tokens | 24 | Jun 4, 2026 | ||
| Instruction Following | FollowBench (test) | 24 | Jun 4, 2026 | ||
| Instruction Following | ComplexBench (test) | 24 | Jun 4, 2026 | ||
| Mathematical Reasoning | MATH 500 | 24 | Jun 4, 2026 | ||
| Node Explanation | MAG | 24 | Jun 4, 2026 | ||
| Node Explanation | ACM | 24 | Jun 4, 2026 | ||
| Line Retrieval | Line Retrieval 600 lines | 24 | Jun 4, 2026 | ||
| Line Retrieval | Line Retrieval 500 lines | 24 | Jun 4, 2026 | ||
| Line Retrieval | Line Retrieval 400 lines | 24 | Jun 4, 2026 | ||
| Line Retrieval | Line Retrieval 300 lines | 24 | Jun 4, 2026 | ||
| Line Retrieval | Line Retrieval 200 lines | 24 | Jun 4, 2026 |