Benchmark
C-Eval
C-Eval is a comprehensive Chinese evaluation suite designed to assess advanced knowledge and reasoning abilities of foundation models in a Chinese…
- Modality
- text
- Categories
- reasoning, general
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 18
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| DeepSeek-V3 | DeepSeek | 0.865 | 2024-12-25 | evidence |
| ERNIE 4.5 | Baidu | 0.407 | 2025-06-25 | evidence |
| Kimi K2 Base | Moonshot AI | 0.925 | 2025-07-11 | evidence |
| Kimi-k1.5 | Moonshot AI | 0.883 | 2025-01-20 | evidence |
| MiMo-V2.5-Pro | Xiaomi | 0.915 | 2026-04-27 | evidence |
| Qwen2 72B Instruct | Qwen | 0.838 | 2024-07-23 | evidence |
| Qwen2 7B Instruct | Qwen | 0.772 | 2024-07-23 | evidence |
| Qwen3.5-0.8B | Qwen | 0.505 | 2026-03-02 | evidence |
| Qwen3.5-122B-A10B | Qwen | 0.919 | 2026-02-24 | evidence |
| Qwen3.5-27B | Qwen | 0.905 | 2026-02-24 | evidence |
| Qwen3.5-2B | Qwen | 0.732 | 2026-03-02 | evidence |
| Qwen3.5-35B-A3B | Qwen | 0.902 | 2026-02-24 | evidence |
| Qwen3.5-397B-A17B | Qwen | 0.93 | 2026-02-16 | evidence |
| Qwen3.5-4B | Qwen | 0.851 | 2026-03-02 | evidence |
| Qwen3.5-9B | Qwen | 0.882 | 2026-03-02 | evidence |
| Qwen3.6 Plus | Qwen | 0.933 | 2026-04-02 | evidence |
| Qwen3.6-27B | Qwen | 0.914 | 2026-04-21 | evidence |
| Qwen3.6-35B-A3B | Qwen | 0.9 | 2026-04-16 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.