OpenAI GPT-6 Astra System Card - U18 - Eating Disorders: leaderboard
Table 2 (U18 evaluations) of OpenAI's GPT-6 Astra System Card (Deployment Safety Hub, 2026-09-03): adherence to teen-specific safety standards on difficult eating disorders cases, assessed with U18 protections in place. Percent (the card prints 0-1), higher is better. Long-tail cases, not estimates of production frequency. GPT-6 Sol, GPT-6 Luna and GPT-5.6 Luna come from Table 24 of the card's section 11 (GPT-6 Sol and GPT-6 Luna, added 2026-09-22), which re-prints the other cells.
Metric: Score (%). Source: deploymentsafety.openai.com. Source: deploymentsafety.openai.com. Source: deploymentsafety.openai.com. Saturation forecast: Around February 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol | 96.2 | #7 |
| 2 | GPT-6.1 Terra | 93.8 | #15 |
| 3 | GPT-6 Astra | 92.1 | #6 |
| 4 | GPT-6 Luna | 87.1 | #61 |
| 5 | GPT-6 Sol | 85.3 | #19 |
| 6 | GPT-5.6 Sol | 71 | #22 |
| 7 | GPT-5.6 Luna | 70.2 | #79 |
| 8 | GPT-5.4 (Thinking) | 67.3 | #83 (GPT-5.4) |
| 9 | GPT-5.5 | 63.9 | #33 |
No result here: #3 Claude Opus 5.5, #9 Gemini 4 Argon, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-6-astra-system-card-u18-eating-disorders · How It Works · Data refreshed daily, snapshot 2026-11-05.