LLM Stats (CSimpleQA): leaderboard
Chinese SimpleQA: short-form factual questions in Chinese testing factual accuracy and knowledge recall on Chinese-centric topics.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Pro (Max) | 84.4 | #102 (DeepSeek V4 Pro) |
| 2 | Qwen 3 235B A22B 2507 Instruct | 84.3 | #300 |
| 3 | Qwen 3 VL 235B A22B Instruct | 83.4 | #273 |
| 4 | DeepSeek V4 Flash (Max) | 78.9 | #137 (DeepSeek V4 Flash) |
| 5 | Kimi K2 | 78.4 | #243 |
| 6 | DeepSeek V3 | 64.8 | #321 |
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-csimpleqa · How It Works · Data refreshed daily, snapshot 2026-10-20.