C-Eval — Results
| Rank | Model | Score |
|---|---|---|
| 1 | qwen3-6-plus | 0.933 |
| 2 | qwen3-5-397b-a17b | 0.93 |
| 3 | kimi-k2-base | 0.925 |
| 4 | qwen3-5-122b-a10b | 0.919 |
| 5 | mimo-v2-5-pro | 0.915 |
| 6 | qwen3-6-27b | 0.914 |
| 7 | qwen3-5-27b | 0.905 |
| 8 | qwen3-5-35b-a3b | 0.902 |
| 9 | qwen3-6-35b-a3b | 0.9 |
| 10 | kimi-k1-5 | 0.883 |
| 11 | qwen3-5-9b | 0.882 |
| 12 | deepseek-v3 | 0.865 |
| 13 | qwen3-5-4b | 0.851 |
| 14 | qwen2-72b-instruct | 0.838 |
| 15 | qwen2-7b-instruct | 0.772 |
| 16 | qwen3-5-2b | 0.732 |
| 17 | qwen3-5-0-8b | 0.505 |
| 18 | ernie-4-5 | 0.407 |
C
C-Eval
1 phaseActive
Comprehensive Chinese evaluation suite with 13,948 MCQs across 52 disciplines at four difficulty levels: middle school, high school, college, and professional.
Overview
C-Eval
Quick answer: C-Eval (Huang et al., 2023) is a comprehensive Chinese evaluation suite with 13,948 multiple-choice questions across 52 disciplines at four difficulty levels. It assesses advanced Chinese knowledge and reasoning across humanities, science, and engineering.
Benchmark Details
| Property | Value |
|---|---|
| Tasks | 13,948 multiple-choice questions |
| Disciplines | 52 (humanities, science, engineering) |
| Difficulty levels | Middle school, high school, college, professional |
| Language | Chinese |
| Metric | Accuracy |
| Subset | C-Eval Hard (very challenging subjects) |
Source: Huang et al. 2023. Last updated 2026-07-24.