Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, Deepseek Deepseek R1 is the more secure of the two: Deepseek V3 scores 47.2% and Deepseek R1 scores 42.6% on attack success rate (ASR) (lower is better).
Deepseek R1 is the overall winner in this comparison!
ASR for Deepseek Deepseek V3 vs Deepseek Deepseek R1. Green marks the safer model on each metric.
Outward is better on every axis.
On Guardion's LLM vulnerability Benchmark, Deepseek Deepseek R1 is the more secure of the two: Deepseek V3 scores 47.2% and Deepseek R1 scores 42.6% on attack success rate (ASR) (lower is better).
Deepseek V3 has a 47.2% 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.
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.