OpenAI Daybreak Red

Daybreak Red is the second alias of OpenAI's Daybreak programme, and the sharper one. Its default snapshot is gpt-5.6-cyber, a model trained for security work rather than adapted to it, and the specification follows: a 400,000-token window with up to 272,000 tokens of input and 128,000 of output, reasoning-token support, text and image input, knowledge to 16 February 2026. OpenAI describes the intended users precisely — approved defenders conducting advanced, authorised vulnerability research, exploit validation and security testing. Every one of those words carries weight, because the same capability read the other way is an attack tool. That is the reason the model is not simply listed for sale: it reaches customers through vetting, and the model card publishes no price. The pairing with Daybreak Blue is deliberate. Blue is the general-purpose flagship with defensive safeguards; Red is a specialist with a narrower window and a narrower permitted use. Both are aliases rather than new weights, which means the interesting question about Daybreak is not what the models can do, but who OpenAI decides may ask them.

▸Videos

No videos yet.

▸Reviews