Benchmark

Natural2Code

reasoning general text

NaturalCodeBench (NCB) is a challenging code benchmark designed to mirror the complexity and variety of real-world coding tasks. It comprises 402 high-quality problems in Python and Java, selected from natural user queries from online coding services, covering 6 different domains.

语言EN
满分1
参评模型8

模型排名

名次 模型 机构 分数 来源
1 Gemini 2.0 Flash Google 92.9 来源 ↗
2 Gemini 1.5 Pro Google 85.4 来源 ↗
3 Gemma 3 27B Google 84.5 来源 ↗
4 Gemma 3 12B Google 80.7 来源 ↗
5 Gemini 1.5 Flash Google 79.8 来源 ↗
6 Gemini 1.5 Flash 8B Google 75.5 来源 ↗
7 Gemma 3 4B Google 70.3 来源 ↗
8 Gemma 3 1B Google 56.0 来源 ↗