language benchmark
CMMLU
CMMLU (Chinese Massive Multitask Language Understanding) is a comprehensive Chinese benchmark that evaluates the knowledge and reasoning capabilities of large language models across 67 different subject topics. The benchmark covers natural sciences, social sciences, engineering, and humanities with multiple-choice questions ranging from basic to advanced professional levels.
Updated Aug 11, 2026
Models6
Model coverage6
MetricScore
EvidenceB
CMMLU Ranking
Higher score ranks better on this benchmark.
| 01 | XI | 90.2% | 100.0% | 6 | C | |
| 02 | AC | 90.1% | 80.0% | 6 | C | |
| 03 | ME | 84.3% | 60.0% | 6 | C | |
| 04 | ME | 82.5% | 40.0% | 6 | C | |
| 05 | OP | 81.5% | 20.0% | 6 | C | |
| 06 | BA | 39.8% | 0.0% | 6 | C |
CMMLU Score Distribution
A closer view of the leading scores on this benchmark.
CMMLU Highlights
The leading models and scores on this benchmark.
What is CMMLU?
What CMMLU measures and how its scores work.
CMMLU (Chinese Massive Multitask Language Understanding) is a comprehensive Chinese benchmark that evaluates the knowledge and reasoning capabilities of large language models across 67 different subject topics. The benchmark covers natural sciences, social sciences, engineering, and humanities with multiple-choice questions ranging from basic to advanced professional levels.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- CMMLU
- Modality
- text
- Primary category
- language
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- cmmlu|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about CMMLU.
Which model scores highest on CMMLU?
MiMo-V2.5-Pro is currently ranked first with 90.2%.
What does CMMLU measure?
CMMLU (Chinese Massive Multitask Language Understanding) is a comprehensive Chinese benchmark that evaluates the knowledge and reasoning capabilities of large language models across 67 different subject topics. The benchmark covers natural sciences, social sciences, engineering, and humanities with multiple-choice questions ranging from basic to advanced professional levels.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
6 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.