OpenAI releases GPT-5.6 to limited US government partners

OpenAI has launched its next-generation GPT-5.6 artificial intelligence model series but is restricting its initial release to a small group of U.S. government-approved partners, the company announced on Friday, June 26, 2026. The staggered rollout follows a direct request from the Trump administration, marking a significant shift in how frontier AI systems are being deployed amid heightened national security concerns.

The GPT-5.6 family consists of three models: Sol, a flagship model for advanced tasks; Terra, a balanced model for everyday applications; and Luna, a faster, lower-cost option. OpenAI stated that Sol delivers improvements in coding, scientific reasoning, biology, cybersecurity, and long-running autonomous tasks. It introduces new operating modes, including a "max" reasoning effort for deeper analysis and an "ultra" mode that uses subagents to orchestrate complex workflows.

Access, however, is tightly controlled. At the request of federal officials, OpenAI is beginning with a limited preview for a select group of trusted partners whose participation has been shared with the government. Reports indicate roughly 20 organizations have received initial access, with the...