OpenAI GPT-5.1 Codex vs Google Gemini 3 Flash
Detailed comparison for LLMs
On Guardion's LLM vulnerability Benchmark, OpenAI GPT-5.1 Codex is the more secure of the two: GPT-5.1 Codex scores 8.5% and Gemini 3 Flash scores 10.0% 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
GPT-5.1 Codex is the overall winner in this comparison!
Attack Success Rate (lower is safer)
ASR for OpenAI GPT-5.1 Codex vs Google Gemini 3 Flash. 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-5.1 Codex has a lower Overall (ASR).
Security Profile
Outward is better on every axis.
Frequently asked questions
Is OpenAI GPT-5.1 Codex or Google Gemini 3 Flash more secure?
On Guardion's LLM vulnerability Benchmark, OpenAI GPT-5.1 Codex is the more secure of the two: GPT-5.1 Codex scores 8.5% and Gemini 3 Flash scores 10.0% 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 GPT-5.1 Codex vs Gemini 3 Flash?
GPT-5.1 Codex has a 8.5% ASR and Gemini 3 Flash has a 10.0% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.
How were GPT-5.1 Codex and Gemini 3 Flash 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.