OpenAI GPT-6 Astra Launch - SRE-Bench: leaderboard
SRE-Bench row of the cybersecurity table in OpenAI's GPT-6 Astra launch post (2026-09-03): models reverse engineer software binaries to recover their core logic without source code. pass@1 in percent (tasks solved in a single attempt), higher is better. Compares GPT-6 Astra, GPT-5.6 Sol and Claude Opus 5.
Metric: pass@1 (%). Source: openai.com. Saturation forecast: Around February 2027. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 88 | #6 |
| 2 | GPT-5.6 Sol | 55.9 | #22 |
| 3 | Claude Opus 5 | 12.5 | #16 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=openai-gpt-6-astra-launch-sre-bench · How It Works · Data refreshed daily, snapshot 2026-11-05.