language benchmark
CSimpleQA
Chinese SimpleQA is the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions. It contains 3,000 high-quality questions spanning 6 major topics with 99 diverse subtopics, designed to assess Chinese factual knowledge across humanities, science, engineering, culture, and society.
Updated Aug 7, 2026
Published models7
Registry coverage7
MetricScore
EvidenceB
CSimpleQA leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | DE | 84.4% | 100.0% | 7 | C | |
| 2 | AC | 84.3% | 83.3% | 7 | C | |
| 3 | AC | 83.4% | 66.7% | 7 | C | |
| 4 | DE | 78.9% | 50.0% | 7 | C | |
| 5 | MA | 78.4% | 33.3% | 7 | C | |
| 6 | MA | 77.6% | 16.7% | 7 | C | |
| 7 | DE | 64.8% | 0.0% | 7 | C |
Score distribution
Top published rows on the benchmark's original scale.
CSimpleQA highlights
The top published results on this benchmark's own scale.
What is CSimpleQA?
Definition and scoring fields from the benchmark registry.
Chinese SimpleQA is the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions. It contains 3,000 high-quality questions spanning 6 major topics with 99 diverse subtopics, designed to assess Chinese factual knowledge across humanities, science, engineering, culture, and society.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- CSimpleQA
- Modality
- text
- Primary category
- language
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- csimpleqa|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about CSimpleQA.
Which model scores highest on CSimpleQA?
DeepSeek-V4-Pro-Max is currently ranked first with 84.4%.
What does CSimpleQA measure?
Chinese SimpleQA is the first comprehensive Chinese benchmark to evaluate the factuality ability of language models to answer short questions. It contains 3,000 high-quality questions spanning 6 major topics with 99 diverse subtopics, designed to assess Chinese factual knowledge across humanities, science, engineering, culture, and society.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
7 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.