OpenAI GPT-5.6 System Card - Prompt Injection - Search and Function Calling: leaderboard

Metric: Prompt-injection robustness rate (%). Source: deploymentsafety.openai.com. 6 models tracked.

Top models

#ModelScore
1GPT-5.6 Terra94.6
2GPT-5.6 Sol91
3GPT-5.6 Luna89.7
4GPT-5.4 (Thinking)69.7
5GPT-5.2 (Thinking)56.8
6GPT-5.1 (Thinking)42.3

Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-5-6-system-card-prompt-injection-search-and-function-calling · How It Works · Data refreshed daily, snapshot 2026-09-19.