Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 557 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Reward Modeling | JudgeBench (test) | 40 | Feb 26, 2026 | ||
| Open ended generation | CNN DailyMail | 40 | Feb 26, 2026 | ||
| Instruction Following | BBH | 40 | Apr 21, 2026 | ||
| Emotional Support Conversation | ESConv and ExTES (val) | 40 | Feb 26, 2026 | ||
| Table Question Answering | TableBench | 40 | Feb 26, 2026 | ||
| Table Question Answering | Penguins in a Table | 40 | Feb 26, 2026 | ||
| Mathematical Problem Solving | MATH | 40 | Apr 21, 2026 | ||
| GUI planning | AndroidControl Low | 40 | Apr 27, 2026 | ||
| Mathematical Reasoning | MATH 500 | 40 | Feb 26, 2026 | ||
| General Question Answering | General QA NQ, TriviaQA, PopQA | 40 | May 8, 2026 | ||
| Multi-round conversation | MT-Bench | 40 | Feb 26, 2026 | ||
| Factuality-based Question Answering | FreshQA 2025/11/24 | 40 | Feb 26, 2026 | ||
| Question Answering | TriviaQA | 40 | Feb 26, 2026 | ||
| Agent Task | AlfWorld | 40 | Jul 7, 2026 | ||
| Safety Evaluation | Harmful Prompts | 40 | Feb 26, 2026 | ||
| Mathematical Reasoning | GSM8K (test) | 40 | Feb 26, 2026 | ||
| Question Answering | TruthfulQA (test) | 40 | Jun 30, 2026 | ||
| Text-to-Image Generation | Text-to-Image Prompts | 40 | May 13, 2026 | ||
| General visual question answering | SimpleVQA | 40 | Jul 2, 2026 | ||
| Person Re-identification | AG-ReID A -> W v2 | 40 | May 19, 2026 | ||
| Backdoor Attack Stealthiness Evaluation | CIFAR10 | 40 | Apr 1, 2026 | ||
| Video Question Answering | VideoMME Overall | 40 | Jul 2, 2026 | ||
| Image Deraining | RealRain1K L | 40 | May 12, 2026 | ||
| Untargeted white-box adversarial attack | ImageNet | 40 | Feb 26, 2026 | ||
| Anomaly Detection | KSDD | 40 | Feb 27, 2026 |