OpenAI o3-mini vs OpenAI GPT-4o
Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, OpenAI o3-mini is the more secure of the two: o3-mini scores 9.0% and GPT-4o scores 22.8% 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
o3-mini is the overall winner in this comparison!
Attack Success Rate (lower is safer)
ASR for OpenAI o3-mini vs OpenAI GPT-4o. 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 o3-mini has a lower Overall (ASR).
- OpenAI GPT-4o has a lower TAP Attack Method (ASR).
- OpenAI GPT-4o has a lower Crescendo Attack Method (ASR).
- OpenAI GPT-4o has a lower Zero-Shot (ASR).
Security Profile
Outward is better on every axis.
Frequently asked questions
Is OpenAI o3-mini or OpenAI GPT-4o more secure?
On Guardion's LLM vulnerability Benchmark, OpenAI o3-mini is the more secure of the two: o3-mini scores 9.0% and GPT-4o scores 22.8% 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 o3-mini vs GPT-4o?
o3-mini has a 9.0% ASR and GPT-4o has a 22.8% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.
How were o3-mini and GPT-4o 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.