Benchmark
CodeForces
math
reasoning
text
A competitive programming benchmark using problems from the CodeForces platform. The benchmark evaluates code generation capabilities of LLMs on algorithmic problems with difficulty ratings ranging from 800 to 2400. Problems cover diverse algorithmic categories including dynamic programming, graph algorithms, data structures, and mathematical problems with standardized evaluation through direct platform submission.
语言EN
满分3000
参评模型6
模型排名
| 名次 | 模型 | 机构 | 分数 | 来源 |
|---|---|---|---|---|
| 1 | GPT OSS 120B | OpenAI | 87.4 | 来源 ↗ |
| 2 | GPT OSS 20B | OpenAI | 83.9 | 来源 ↗ |
| 3 | DeepSeek-V3.2-Exp | DeepSeek | 70.7 | 来源 ↗ |
| 4 | DeepSeek-V3.1 | DeepSeek | 69.7 | 来源 ↗ |
| 5 | Qwen3 32B | Alibaba Cloud / Qwen Team | 65.9 | 来源 ↗ |
| 6 | DeepSeek-R1-0528 | DeepSeek | 64.3 | 来源 ↗ |