LLM Stats (CMMLU): leaderboard
Chinese Massive Multitask Language Understanding: 11,528 questions across 67 subjects testing knowledge in Chinese-specific domains including law, medicine, and traditional culture.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MiMo-V2.5-Pro | 90.2 | #158 |
| 2 | Qwen 2 72B Instruct | 90.1 | #576 |
| 3 | LongCat-Flash-Chat | 84.34 | #331 |
| 4 | LongCat Flash Lite | 82.48 | #681 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-cmmlu · How It Works · Data refreshed daily, snapshot 2026-10-20.