OpenAI GPT-5.6 August Update System Card - Image Input - Self-Harm: leaderboard
Metric: not_unsafe rate (%). Source: deploymentsafety.openai.com. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.3 Instant | 87.1 |
| 2 | GPT-5.5 Instant | 86.8 |
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-5-6-august-update-system-card-image-input-self-harm · How It Works · Data refreshed daily, snapshot 2026-09-19.