Google Gemini 2.5 Pro vs MiniMax MiniMax M2

Detailed comparison for LLMs

GoogleMiniMax

On Guardion's LLM vulnerability Benchmark, Google Gemini 2.5 Pro is the more secure of the two: Gemini 2.5 Pro scores 16.1% and MiniMax M2 scores 40.0% 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

Gemini 2.5 Pro is the overall winner in this comparison!

Attack Success Rate (lower is safer)

ASR for Google Gemini 2.5 Pro vs MiniMax MiniMax M2. Green marks the safer model on each metric. Only the overall score is available for estimated models.

Overall (ASR)

Gemini 2.5 Pro
16.1%
MiniMax M2
40.0%

TAP Attack Method (ASR)

Gemini 2.5 Pro
27.5%
MiniMax M2
100.0%

Crescendo Attack Method (ASR)

Gemini 2.5 Pro
19.1%
MiniMax M2
100.0%

Zero-Shot (ASR)

Gemini 2.5 Pro
1.6%
MiniMax M2
100.0%

Key Highlights

  • Google Gemini 2.5 Pro has a lower Overall (ASR).
  • Google Gemini 2.5 Pro has a lower TAP Attack Method (ASR).
  • Google Gemini 2.5 Pro has a lower Crescendo Attack Method (ASR).
  • Google Gemini 2.5 Pro has a lower Zero-Shot (ASR).

Security Profile

Outward is better on every axis.

OverallTAPCrescendoZero-Shot
Gemini 2.5 Pro
MiniMax M2
Full security profile
Google Gemini 2.5 Pro
Full security profile
MiniMax MiniMax M2

Frequently asked questions

Is Google Gemini 2.5 Pro or MiniMax MiniMax M2 more secure?

On Guardion's LLM vulnerability Benchmark, Google Gemini 2.5 Pro is the more secure of the two: Gemini 2.5 Pro scores 16.1% and MiniMax M2 scores 40.0% 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 Gemini 2.5 Pro vs MiniMax M2?

Gemini 2.5 Pro has a 16.1% ASR and MiniMax M2 has a 40.0% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.

How were Gemini 2.5 Pro and MiniMax M2 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