Anthropic Claude Sonnet 4.5 vs OpenAI GPT-5.2 Pro
Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, OpenAI GPT-5.2 Pro is the more secure of the two: Claude Sonnet 4.5 scores 4.2% and GPT-5.2 Pro scores 3.8% 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
GPT-5.2 Pro is the overall winner in this comparison!
Attack Success Rate (lower is safer)
ASR for Anthropic Claude Sonnet 4.5 vs OpenAI GPT-5.2 Pro. Green marks the safer model on each metric. Only the overall score is available for estimated models.
Overall (ASR)
TAP Attack Method (ASR)
Crescendo Attack Method (ASR)
Zero-Shot (ASR)
Key Highlights
- OpenAI GPT-5.2 Pro has a lower Overall (ASR).
Security Profile
Outward is better on every axis.
Frequently asked questions
Is Anthropic Claude Sonnet 4.5 or OpenAI GPT-5.2 Pro more secure?
On Guardion's LLM vulnerability Benchmark, OpenAI GPT-5.2 Pro is the more secure of the two: Claude Sonnet 4.5 scores 4.2% and GPT-5.2 Pro scores 3.8% 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 Claude Sonnet 4.5 vs GPT-5.2 Pro?
Claude Sonnet 4.5 has a 4.2% ASR and GPT-5.2 Pro has a 3.8% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.
How were Claude Sonnet 4.5 and GPT-5.2 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.