Alibaba Qwen2 7B Instruct vs Mistral Mistral NeMo Instruct

Detailed comparison for LLMs

AlibabaMistral

On Guardion's LLM vulnerability Benchmark, Mistral Mistral NeMo Instruct is the more secure of the two: Qwen2 7B Instruct scores 52.0% and Mistral NeMo Instruct scores 50.0% 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

Mistral NeMo Instruct is the overall winner in this comparison!

Attack Success Rate (lower is safer)

ASR for Alibaba Qwen2 7B Instruct vs Mistral Mistral NeMo Instruct. Green marks the safer model on each metric. Only the overall score is available for estimated models.

Overall (ASR)

Qwen2 7B Instruct
52.0%
Mistral NeMo Instruct
50.0%

TAP Attack Method (ASR)

Qwen2 7B Instruct
100.0%
Mistral NeMo Instruct
100.0%

Crescendo Attack Method (ASR)

Qwen2 7B Instruct
100.0%
Mistral NeMo Instruct
100.0%

Zero-Shot (ASR)

Qwen2 7B Instruct
100.0%
Mistral NeMo Instruct
100.0%

Key Highlights

  • Mistral Mistral NeMo Instruct has a lower Overall (ASR).

Security Profile

Outward is better on every axis.

OverallTAPCrescendoZero-Shot
Qwen2 7B Instruct
Mistral NeMo Instruct
Full security profile
Alibaba Qwen2 7B Instruct
Full security profile
Mistral Mistral NeMo Instruct

Frequently asked questions

Is Alibaba Qwen2 7B Instruct or Mistral Mistral NeMo Instruct more secure?

On Guardion's LLM vulnerability Benchmark, Mistral Mistral NeMo Instruct is the more secure of the two: Qwen2 7B Instruct scores 52.0% and Mistral NeMo Instruct scores 50.0% 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 Qwen2 7B Instruct vs Mistral NeMo Instruct?

Qwen2 7B Instruct has a 52.0% ASR and Mistral NeMo Instruct has a 50.0% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.

How were Qwen2 7B Instruct and Mistral NeMo Instruct 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.

Related Comparisons