Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 849 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Reasoning | BBH, GSM8K | 30 | Apr 14, 2026 | ||
| Mathematical Reasoning | AIME25 | 30 | Apr 14, 2026 | ||
| Mathematical Reasoning | AIME24 | 30 | Apr 14, 2026 | ||
| Mathematical Reasoning | MATH500 | 30 | Apr 14, 2026 | ||
| Mathematical Reasoning | GSM8K | 30 | Apr 14, 2026 | ||
| Knowledge corruption traceback | NQ | 30 | Apr 14, 2026 | ||
| Cross-lingual Aspect-Based Sentiment Analysis | SemEval 2016 | 30 | Apr 13, 2026 | ||
| Vision-dependent Multimodal Reasoning | Vision-dependent Multimodal Reasoning Suite (LogicVista, Counting, MMMU-Pro, MathVerseV) | 30 | Jun 8, 2026 | ||
| Multi-view Classification | UCI | 30 | Jun 5, 2026 | ||
| Refusal Rate Evaluation | TextVQA | 30 | Apr 13, 2026 | ||
| Refusal Rate Evaluation | CelebA | 30 | Apr 13, 2026 | ||
| Refusal Rate Evaluation | GQA | 30 | Apr 13, 2026 | ||
| Refusal Rate Evaluation | VQA v2 | 30 | Apr 13, 2026 | ||
| Retrieval | LongDocURL | 30 | Jun 9, 2026 | ||
| Vision Question Answering | OKVQA | 30 | Apr 10, 2026 | ||
| Planning | Openscene (val) | 30 | Apr 10, 2026 | ||
| Data Selection | Navtrain (val) | 30 | Apr 10, 2026 | ||
| Response Generation | Chronicle and MSC Average | 30 | Apr 10, 2026 | ||
| Zero-shot evaluation | Downstream Tasks PiQA ARC Hellaswag Winogrande BoolQ | 30 | Apr 10, 2026 | ||
| Mathematical Reasoning | AIME 2025 | 30 | Apr 10, 2026 | ||
| Mathematical Reasoning | GSM8K | 30 | Jun 16, 2026 | ||
| Question Answering | NQ | 30 | Apr 10, 2026 | ||
| Question Answering | MS-MARCO | 30 | Apr 10, 2026 | ||
| Question Answering | HotpotQA | 30 | Apr 10, 2026 | ||
| Regression | ATACH | 30 | Apr 10, 2026 |