Knowledge · Benchmark profile
A Chinese short-form factuality benchmark reported by DeepSeek for V4 model evaluations.
Data verified 27 Jul 2026 · Methodology 1.8.0
Benchmark score on Chinese-SimpleQA
DeepSeek V4 Pro (Max) leads at 84.4%, followed by DeepSeek V4 Flash (Max) (78.9%) and DeepSeek V4 Pro (High) (77.7%).
6 ranked models · higher is better · labels show rank and score
| Model | Rank | Provider | Score |
|---|---|---|---|
| DeepSeek V4 Pro (Max) | #1 | DeepSeek | 84.4% |
| DeepSeek V4 Flash (Max) | #2 | DeepSeek | 78.9% |
| DeepSeek V4 Pro (High) | #3 | DeepSeek | 77.7% |
| DeepSeek V4 Pro | #4 | DeepSeek | 75.8% |
| DeepSeek V4 Flash (High) | #5 | DeepSeek | 73.2% |
| DeepSeek V4 Flash | #6 | DeepSeek | 71.5% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | DeepSeek V4 Pro (Max) deepseek-v4-pro-max | DeepSeek | open | Estimated reference | 84.4% |
| #2 | DeepSeek V4 Flash (Max) deepseek-v4-flash-max | DeepSeek | open | Estimated reference | 78.9% |
| #3 | DeepSeek V4 Pro (High) deepseek-v4-pro-high | DeepSeek | open | Estimated reference | 77.7% |
| #4 | DeepSeek V4 Pro deepseek-v4-pro | DeepSeek | open | Reference only | 75.8% |
| #5 | DeepSeek V4 Flash (High) deepseek-v4-flash-high | DeepSeek | open | Estimated reference | 73.2% |
| #6 | DeepSeek V4 Flash deepseek-v4-flash | DeepSeek | open | Estimated reference | 71.5% |
The top of this snapshot is led by DeepSeek V4 Pro (Max) at 84.4%; third place is 6.7 points behind. The top-6 spread is 12.9 points.
About Chinese-SimpleQA
A Chinese short-form factuality benchmark reported by DeepSeek for V4 model evaluations. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A Chinese short-form factuality benchmark reported by DeepSeek for V4 model evaluations.
DeepSeek V4 Pro (Max) by DeepSeek currently leads with 84.4%.
6 models in the LuminaBench cohort have a qualifying score on this benchmark.
Related
2026 · 21 results · Reference
Artificial Analysis MMLU-Pro2026 · 8 results · Reference
Massive Multitask Language Understanding2020 · 16 results · Reference
Graduate-Level Google-Proof Q&A2023 · 140 results · Reference
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines2025 · 38 results · Reference
Massive Multitask Language Understanding Professional2024 · 85 results · Reference
