OpenAI GPT-6.1 Terra System Card - GPT-Red Indirect Prompt Injection: leaderboard
Defender success rate (%; averaged per query); GPT-6.1 Terra October 26, 2026 system-card cohort; native printed labels, with evaluation and configuration boundaries in board/model names. Metric: Defender success rate (%; averaged per query).
Metric: Defender success rate (%; averaged per query). Source: deploymentsafety.openai.com. Saturation forecast: Estimated already saturated. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 99.79 | #5 |
| 2 | GPT-6 Sol | 99.05 | #17 |
| 3 | GPT-6.1 Sol | 98.98 | #6 |
| 4 | GPT-6.1 Terra | 98.77 | #10 |
| 5 | GPT-6 Luna | 98.61 | #60 |
No result here: #3 Claude Opus 5.5, #8 Gemini 4 Argon, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-6-1-terra-system-card-gpt-red-indirect-prompt-injection · How It Works · Data refreshed daily, snapshot 2026-10-27.