OpenAI has announced that it will publicly launch all three variants of its new GPT-5.6 model series — named Sol, Luna and Terra — on Thursday 9 July, after receiving government clearance for a wider release. The launch matters because it is one of the first tests of a new US framework in which AI firms voluntarily submit their most powerful models for federal review before public release, following an AI cybersecurity order signed by President Trump in early June.

The models were initially limited in late June to a small group of trusted partners and government-approved entities, pending review. The Department of Commerce's Center for AI Standards and Innovation carried out additional testing, and OpenAI sent technical experts to Washington before approval was granted in well under the 30-day period the order allows. Sol is described as OpenAI's most powerful model to date, Terra as a cheaper everyday option matching GPT-5.5's performance, and Luna as its lowest-cost model, with per-token prices ranging accordingly. Rival Anthropic faced similar constraints, having had to block access to its Mythos and Fable models before regaining permission to redeploy them.