Anthropic Claude 3.7 Sonnet vs Google Gemini 2.5 Pro
Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, Google Gemini 2.5 Pro is the more secure of the two: Claude 3.7 Sonnet scores 20.3% and Gemini 2.5 Pro scores 16.1% on attack success rate (ASR) (lower is better).
Head-to-Head Overview
Gemini 2.5 Pro is the overall winner in this comparison!
Attack Success Rate (lower is safer)
ASR for Anthropic Claude 3.7 Sonnet vs Google Gemini 2.5 Pro. Green marks the safer model on each metric.
Overall (ASR)
TAP Attack Method (ASR)
Crescendo Attack Method (ASR)
Zero-Shot (ASR)
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).
Security Profile
Outward is better on every axis.
Frequently asked questions
Is Anthropic Claude 3.7 Sonnet or Google Gemini 2.5 Pro more secure?
On Guardion's LLM vulnerability Benchmark, Google Gemini 2.5 Pro is the more secure of the two: Claude 3.7 Sonnet scores 20.3% and Gemini 2.5 Pro scores 16.1% on attack success rate (ASR) (lower is better).
What is the attack success rate (ASR) of Claude 3.7 Sonnet vs Gemini 2.5 Pro?
Claude 3.7 Sonnet has a 20.3% ASR and Gemini 2.5 Pro has a 16.1% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.
How were Claude 3.7 Sonnet and Gemini 2.5 Pro 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.