// BENCHMARK

C-SimpleQA benchmark

AI model leaderboard for the C-SimpleQA benchmark. Compare how large language models score on C-SimpleQA, see the full ranking, and understand what this AI benchmark measures. DeepSeek-V3 currently leads with 68.9%. Chinese factuality benchmark — Chinese-language counterpart of SimpleQA.

Leaderboard

# Model Organization Score Variant Source
#1 DeepSeek-V3 DeepSeek 68.9% official ↗
#2 MiniMax-Text-01 MiniMax 67.4% from-hf-readme official ↗
#3 DeepSeek-V3 DeepSeek 64.8% cited-minimax-text official ↗
#4 GPT-4o (Nov 2024) OpenAI 64.6% cited-minimax-text official ↗
#5 Gemini 2.0 Flash Google DeepMind 63.3% exp-cited-minimax-text official ↗
#6 Gemini 1.5 Pro Google DeepMind 59.4% 002-cited-minimax-text official ↗
#7 Claude 3.5 Sonnet Anthropic 56.8% 1022-cited-minimax-text official ↗
#8 dots.llm1 Rednote 56.7% official ↗
#9 Llama 3.1-405B Meta AI 54.7% instruct-cited-minimax-text official ↗
#10 Qwen2.5 Instruct (72B) Qwen 52.2% instruct-cited-minimax-text official ↗

Frequently asked questions about C-SimpleQA

What is the C-SimpleQA benchmark?

Chinese factuality benchmark — Chinese-language counterpart of SimpleQA.

How is the C-SimpleQA benchmark scored?

C-SimpleQA is scored using the accuracy (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.

Which AI model scores highest on C-SimpleQA?

As of the latest reported scores on GenAIList, DeepSeek-V3 achieves the highest result on C-SimpleQA with a score of 68.9%.

Is a higher C-SimpleQA score better?

Yes. On C-SimpleQA a higher score indicates better performance, so models near the top of the leaderboard are the strongest.