OpenAI GPT-5.6 System Card - Challenging Prompts - Hate: leaderboard
Metric: not_unsafe rate (%). Source: deploymentsafety.openai.com. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 100 |
| 2 | GPT-5.6 Luna | 100 |
| 3 | GPT-5.6 Terra | 100 |
| 4 | GPT-5.6 Sol | 98.2 |
| 5 | GPT-5.4 (Thinking) | 94.3 |
| 6 | GPT-5.2 (Thinking) | 92.7 |
| 7 | GPT-5.1 (Thinking) | 80.8 |
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-5-6-system-card-challenging-prompts-hate · How It Works · Data refreshed daily, snapshot 2026-09-19.