LLM Stats (C-Eval) Leaderboard — The Aggregate

LLM Stats (C-Eval) — leaderboard

Metric: Score (%). Source: llm-stats.com. 18 models tracked.

Top models

#	Model	Score
1	Qwen 3.6 Plus	93.3
2	Qwen 3.5 397B A17B	93
3	Qwen 3.5 122B A10B	91.9
4	MiMo-V2.5-Pro	91.5
5	Qwen 3.6 27B	91.4
6	Qwen 3.5 27B	90.5
7	Qwen 3.5 35B A3B	90.2
8	Qwen 3.6 35B A3B	90
9	Qwen 3.5 9B	88.2
10	DeepSeek V3	86.5
11	Qwen 3.5 4B	85.1
12	Qwen 2 72B Instruct	83.8
13	Qwen 2 7B Instruct	77.2
14	Qwen 3.5 2B	73.2
15	Qwen 3.5 0.8B	50.5

Interactive version: theaggregate.ai/benchmark?slug=llm-stats-c-eval · How the rankings work · Data refreshed daily, snapshot 2026-07-22.