OpenAI to launch GPT-5.6 series Thursday after US government lifts restrictions
OpenAI announced that its new GPT-5.6 family of AI models, including flagship Sol, will be publicly available on 9 July, following a limited preview for US partners and a reported approval from the Trump administration.
The announcement
OpenAI said on Tuesday that its most advanced model series to date, GPT-5.6, will launch publicly on Thursday. The company posted on X that it is "expanding preview access globally now." The release follows a restricted early-access phase that began at the end of June, when a small group of US-based partners were granted access at Washington's request.
GPT-5.6 Sol launches Thursday! Happy building.
Three tiers
GPT-5.6 comprises three models: Sol, the new flagship; Terra, a mid-range model for everyday work that costs half as much as the previous GPT-5.5; and Luna, a fast, low-cost option. The tiered lineup mirrors a broader industry trend toward offering performance and price flexibility.
Security concerns
The new models, along with Anthropic's Mythos series, have drawn scrutiny for their alleged ability to identify software vulnerabilities that hackers could exploit. OpenAI said Sol had been tested for "several weeks" to strengthen safeguards before the limited preview.
Regulatory backdrop
The US government had earlier this month imposed restrictions on Anthropic's Fable 5 and Mythos 5 models, briefly forcing their withdrawal over a directive that barred use by foreign nationals. Those restrictions were lifted at the end of June, and Anthropic said last week it was restoring global access. Axios reported, citing a source familiar with the matter, that the Trump administration gave OpenAI the green light for a wide release after technical tests and meetings with company officials.
We are now expanding preview access globally.
- Anthropic takes down Fable 5 and Mythos 5 after White House directive bars use by foreign nationals.
- OpenAI grants early access to GPT-5.6 for a limited group of US-based partners at Washington's request.
- US lifts restrictions on Mythos 5, allowing redeployment to cybersecurity specialists and infrastructure providers.
- Anthropic announces it will restore global access to Fable 5 and Mythos 5.
- OpenAI announces on X that GPT-5.6 Sol, Terra, and Luna will launch publicly on Thursday.
- GPT-5.6 models become publicly available worldwide.
Industry ripple effects
The parallel treatment of OpenAI and Anthropic suggests a coordinated US policy shift on frontier AI deployment. Both companies are now moving from tightly controlled rollouts to broad public availability within days of each other.
