OpenAI is restricting the release of its newest artificial intelligence models to a “small group of trusted partners” at the behest of the US government, the company said on Friday.
The next generation of the GPT-5.6 series includes Sol, its flagship model. Terra, a more balanced model for everyday use. and Luna, a faster, lower-cost option. Although the Sol is the company’s most powerful model, the Trump administration has restricted the release of all three. OpenAI said the preview is limited to partners “whose participation has been shared with the government.”
The administration’s request comes as the US government puts new pressure on artificial intelligence companies to scale back their most advanced systems. After Anthropic released the more powerful Fable 5 public model, management ordered the company to remove access for any foreign nationals, resulting in Anthropic removing the model entirely.
The incident has raised questions about how much power the government should have over AI model releases. Dean Ball, a former White House AI adviser and future OpenAI employee, says President Trump’s recent executive order — requiring some AI companies to voluntarily submit their most advanced models for government review up to 30 days before launch — has created a de facto involuntary licensing regime for frontier AI, leading to severe restrictions.
The problem is compounded, Ball argues, when the government lacks clearly defined safety standards, which could lead to endless launch delays that could not only help China in the AI race, but also jeopardize the billions of dollars allocated to building AI infrastructure.
And while OpenAI did as management asked this time, the AI company made it clear it wasn’t happy with the deal.
“We do not believe that this type of government access process should become the long-term default,” a Friday statement said. blog post. “It curates the best tools from users, developers, enterprises, cyber defenders and global partners who need them.”
OpenAI called the preview a “short-term step” that will put GPT-5.6 on a path to wider availability in the coming weeks, as the company works with the administration to develop a new cybersecurity executive order framework, as well as an “iterative process for future model releases.”
Specifications GPT-5.6 Sol
OpenAI says GPT-5.6 Sol is its most powerful model to date, with improved agent capabilities in coding, biology, and cybersecurity. Sol introduces a “max” reasoning mode and an “excessive” mode that uses coordinated subfactors to solve extremely complex tasks (just the kind of neat gimmick that skyrockets your token usage).
GPT-5.6 excels on several benchmarks, OpenAI says, including being slightly better at coding workflow than Anthropic’s Claude Mythos 5, which the Trump administration also effectively banned this month. OpenAI says GPT-5.6 Sol is also competitive with the Mythos preview, but uses a third of the output tokens.
To allay any fears that its powerful models are not secure, OpenAI says that Sol includes its most robust security stack yet. It is, OpenAI says, highly hardened against adversary attacks and deliberately optimized to favor defensive cyber work over offensive exploits. In other words, it’s designed to be difficult to jailbreak while prioritizing showing users how to defend against exploits rather than how to hack systems.
OpenAI also says that guardrails are built directly into the behavior of the base model, rather than relying on a separate filter on top of it. The company is likely trying to avoid the trap that Anthropic fell into with Fable 5. In the brief moments that Fable 5 was available, whenever the model’s classifiers spotted a high-risk topic — like cybersecurity, biology, or chemistry — it wouldn’t just block the message. will direct the request to an older model. The whole overly careful flow and invisible routing led to a lot of false positives and user backlash.
While the GPT-5.6 models are initially only available to a select group of partners, OpenAI plans to make them more widely available to people using ChatGPT, the Codex, and the API soon.
GPT-5.6 comes in three sizes with tiered pricing: Sol costs $5 per million input tokens and $30 per million output tokens. Terra costs half that. and Luna costs $1 and $6, respectively. OpenAI says it has also improved prompt caching to make repeated prompts cheaper and more predictable.
When you purchase through links in our articles, we may earn a small commission. This does not affect our editorial independence.
