DeepSeek DeepSeek-V3.2 Thinking vs Deepseek Deepseek R1
Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, DeepSeek DeepSeek-V3.2 Thinking is the more secure of the two: DeepSeek-V3.2 Thinking scores 42.0% and Deepseek R1 scores 42.6% on attack success rate (ASR) (lower is better). One or both scores are estimated from public safety evaluations pending a Guardion benchmark run.
Head-to-Head Overview
DeepSeek-V3.2 Thinking is the overall winner in this comparison!
Attack Success Rate (lower is safer)
ASR for DeepSeek DeepSeek-V3.2 Thinking vs Deepseek Deepseek R1. Green marks the safer model on each metric. Only the overall score is available for estimated models.
Overall (ASR)
TAP Attack Method (ASR)
Crescendo Attack Method (ASR)
Zero-Shot (ASR)
Key Highlights
- DeepSeek DeepSeek-V3.2 Thinking has a lower Overall (ASR).
- Deepseek Deepseek R1 has a lower TAP Attack Method (ASR).
- Deepseek Deepseek R1 has a lower Crescendo Attack Method (ASR).
- Deepseek Deepseek R1 has a lower Zero-Shot (ASR).
Security Profile
Outward is better on every axis.
Frequently asked questions
Is DeepSeek DeepSeek-V3.2 Thinking or Deepseek Deepseek R1 more secure?
On Guardion's LLM vulnerability Benchmark, DeepSeek DeepSeek-V3.2 Thinking is the more secure of the two: DeepSeek-V3.2 Thinking scores 42.0% and Deepseek R1 scores 42.6% on attack success rate (ASR) (lower is better). One or both scores are estimated from public safety evaluations pending a Guardion benchmark run.
What is the attack success rate (ASR) of DeepSeek-V3.2 Thinking vs Deepseek R1?
DeepSeek-V3.2 Thinking has a 42.0% ASR and Deepseek R1 has a 42.6% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.
How were DeepSeek-V3.2 Thinking and Deepseek R1 tested?
Both were red-teamed with the HarmBench framework across zero-shot, TAP (Tree of Attacks with Pruning), and Crescendo multi-turn attacks, scored by Attack Success Rate.