OpenAI has begun the full public rollout of its ChatGPT 5.6 model family, led by a flagship system called Sol, after weeks of staggered access tied to a voluntary safety review arrangement with the White House. In a blog post announcing the launch, OpenAI said Sol 'launches with our most robust safety stack to date'.
The release, which also includes two smaller models named Terra and Luna, mirrors a standoff earlier this year involving rival firm Anthropic, whose Claude Fable and Mythos models were briefly taken offline under actual US export controls, a legally distinct process from the arrangement OpenAI has been operating under.
