Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, MiniMax MiniMax M1 40K is the more secure of the two: MiniMax M1 40K scores 43.5% and Qwen3 VL 8B Thinking scores 44.0% on attack success rate (ASR) (lower is better). One or both scores are estimated from public safety evaluations pending a Guardion benchmark run.
MiniMax M1 40K is the overall winner in this comparison!
ASR for MiniMax MiniMax M1 40K vs Alibaba Qwen3 VL 8B Thinking. Green marks the safer model on each metric. Only the overall score is available for estimated models.
Outward is better on every axis.
On Guardion's LLM vulnerability Benchmark, MiniMax MiniMax M1 40K is the more secure of the two: MiniMax M1 40K scores 43.5% and Qwen3 VL 8B Thinking scores 44.0% on attack success rate (ASR) (lower is better). One or both scores are estimated from public safety evaluations pending a Guardion benchmark run.
MiniMax M1 40K has a 43.5% ASR and Qwen3 VL 8B Thinking has a 44.0% 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.