OpenAI GPT-5.6 System Card - Prompt Injection - Search and Function Calling: leaderboard
Table 5 (prompt injection) of OpenAI's GPT-5.6 System Card (Deployment Safety Hub, 2026-07-09): robustness to stronger prompt-injection attacks embedded in search and function-calling tool outputs. Percent (the card prints 0-1), higher is better. GPT-5.5 has no value on this row.
Metric: Prompt-injection robustness rate (%). Source: deploymentsafety.openai.com. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Terra | 94.6 | #40 |
| 2 | GPT-5.6 Sol | 91 | #22 |
| 3 | GPT-5.6 Luna | 89.7 | #79 |
| 4 | GPT-5.4 (Thinking) | 69.7 | #83 (GPT-5.4) |
| 5 | GPT-5.2 (Thinking) | 56.8 | #115 (GPT-5.2) |
| 6 | GPT-5.1 (Thinking) | 42.3 | #145 (GPT-5.1) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
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 2026-11-05.