Benchmark
LiveBench
math
reasoning
general
text
LiveBench is a challenging, contamination-limited LLM benchmark that addresses test set contamination by releasing new questions monthly based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses. It comprises tasks across math, coding, reasoning, language, instruction following, and data analysis with verifiable, objective ground-truth answers.
语言EN
满分1
参评模型13
模型排名
| 名次 | 模型 | 机构 | 分数 | 来源 |
|---|---|---|---|---|
| 1 | o3-mini | OpenAI | 84.6 | 来源 ↗ |
| 2 | Qwen3 235B A22B | Alibaba Cloud / Qwen Team | 77.1 | 来源 ↗ |
| 3 | Kimi K2 Instruct | Moonshot AI | 76.4 | 来源 ↗ |
| 4 | Kimi K2-Instruct-0905 | Moonshot AI | 76.4 | 来源 ↗ |
| 5 | Qwen3 32B | Alibaba Cloud / Qwen Team | 74.9 | 来源 ↗ |
| 6 | Qwen3 30B A3B | Alibaba Cloud / Qwen Team | 74.3 | 来源 ↗ |
| 7 | QwQ-32B | Alibaba Cloud / Qwen Team | 73.1 | 来源 ↗ |
| 8 | o1 | OpenAI | 67.0 | 来源 ↗ |
| 9 | o1-preview | OpenAI | 52.3 | 来源 ↗ |
| 10 | Qwen2.5 72B Instruct | Alibaba Cloud / Qwen Team | 52.3 | 来源 ↗ |
| 11 | Phi 4 | Microsoft | 47.6 | 来源 ↗ |
| 12 | Qwen2.5 7B Instruct | Alibaba Cloud / Qwen Team | 35.9 | 来源 ↗ |
| 13 | Qwen2.5-Omni-7B | Alibaba Cloud / Qwen Team | 29.6 | 来源 ↗ |