Google LLC and OpenAI Launch Next-Generation AI Models

Major AI developers expanded their model offerings in July 2026, introducing more efficient and specialized tools for developers and enterprises. OpenAI publicly released the GPT-5.6 model family on July 9, featuring the flagship Sol, balanced Terra, and cost-efficient Luna versions. This broad release followed a 13-day restricted period starting June 26, during which the U.S. government limited access to trusted partners for security testing. This restriction stemmed from a June 2 executive order signed by President Donald Trump to ensure the secure deployment of frontier AI, and the launch only proceeded after approval from the U.S. Department of Commerce.

Shortly after, Google LLC released Gemini 3.6 Flash and Gemini 3.5 Flash-Lite on July 21. The company aimed to reduce costs for developers, lowering output token costs for Gemini 3.6 Flash to $7.50 per million. By July 28, Google LLC established Gemini 3.6 Flash as the default for its Gemini API Managed Agents platform and introduced new production controls, such as token budget caps and scheduled execution triggers.

Both companies introduced professional-grade tools alongside these releases. OpenAI launched ChatGPT Work, while Google LLC unveiled a cybersecurity-focused AI built on the Gemini framework. Google LLC also expanded free tier access for Managed Agents and announced that training for Gemini 4 has already begun.