GPT-5.6-Cyber

OpenAI's cybersecurity-specific model, available only to defenders vetted into the new Daybreak Red access tier.

Built on GPT-5.6 Sol and trained on specialized cybersecurity tasks including finding zero-day vulnerabilities and developing exploit chains
Refuses far less on higher-risk dual-use requests: it completed 95% of advanced test requests covering exploit chains, authentication bypasses, and privilege escalation, against 57.3% for GPT-5.5-Cyber and 1.5% for GPT-5.6 Sol under normal safeguards
OpenAI split Daybreak into two tiers. Daybreak Blue gives approved defenders general-purpose models like GPT-5.6 Sol for vulnerability discovery, secure code review, malware analysis, and incident response. Daybreak Red gates the purpose-trained cyber models behind tighter vetting
Access requires approval into Daybreak Red. There is no self-serve or public option
OpenAI used the model to find two previously unknown vulnerabilities in Chrome's V8 engine that could be chained to escape its sandbox