Alibaba Qwen3 VL 32B Thinking vs DeepSeek DeepSeek VL2
Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, Alibaba Qwen3 VL 32B Thinking is the more secure of the two: Qwen3 VL 32B Thinking scores 38.0% and DeepSeek VL2 scores 52.0% 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
Qwen3 VL 32B Thinking is the overall winner in this comparison!
Attack Success Rate (lower is safer)
ASR for Alibaba Qwen3 VL 32B Thinking vs DeepSeek DeepSeek VL2. 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
- Alibaba Qwen3 VL 32B Thinking has a lower Overall (ASR).
Security Profile
Outward is better on every axis.
Frequently asked questions
Is Alibaba Qwen3 VL 32B Thinking or DeepSeek DeepSeek VL2 more secure?
On Guardion's LLM vulnerability Benchmark, Alibaba Qwen3 VL 32B Thinking is the more secure of the two: Qwen3 VL 32B Thinking scores 38.0% and DeepSeek VL2 scores 52.0% 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 Qwen3 VL 32B Thinking vs DeepSeek VL2?
Qwen3 VL 32B Thinking has a 38.0% ASR and DeepSeek VL2 has a 52.0% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.
How were Qwen3 VL 32B Thinking and DeepSeek VL2 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.