OpenAI and Google, usually rivals racing to out-build each other's chatbots, have struck a deal that puts OpenAI's models inside Google's own cloud platform.
Under the arrangement, OpenAI's frontier GPT-6 series and its Codex coding tools are coming to Google Cloud's Model Garden, a library where businesses and developers pick AI models to build with. Three models, GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna, have started rolling out in private preview to select enterprise customers. OpenAI says more will follow at an unannounced date.
The three models split the work differently. GPT-6 Astra is built for heavier lifting: complex coding, computer-use tasks, professional work and scientific research. GPT-6.1 Sol sits close to Astra in capability and handles similar agentic coding and professional tasks. GPT-6 Luna is the budget option, meant for everyday work applied at scale rather than frontier-level reasoning.
This isn't OpenAI's first move outside Microsoft's cloud. Its models and Codex tools landed on Amazon Bedrock earlier this year, including the GPT-6 and GPT-5.6 lines along with specialized Daybreak cybersecurity models. Microsoft remains OpenAI's primary cloud partner, but the Google deal and the Amazon one before it show OpenAI spreading its distribution well beyond the one company that has bankrolled much of its growth.
The plan doesn't stop at the Model Garden. OpenAI's frontier models are also set to show up in Google Antigravity and the Gemini Agent, Google's own developer and agent tools, putting a competitor's models inside Google's own products.
What this isn't is a consumer-facing change. Nobody using the ChatGPT app or the Gemini app is getting access to the other company's models through this deal. It's aimed squarely at enterprise customers and developers who already work inside Google Cloud and want the option to run OpenAI's models without switching platforms.
Why the companies are framing this around control
In a blog post announcing the partnership, OpenAI leaned on the language of enterprise compliance rather than performance. "For organizations in regulated industries, deploying AI requires control over who can access sensitive data and how it is used," the company wrote, adding that the integration pairs its models with Google Cloud's access controls, data governance tools and infrastructure.
Greg Brockman, OpenAI's president and co-founder, called Google Cloud "a leading platform for developers and businesses building AI agents." Thomas Kurian, Google Cloud's CEO, said the company is "focused on giving customers a broad selection of models, including those from OpenAI, so they can choose the right model for any given job."
That framing matters because Google Cloud's Model Garden already carries more than 200 first-party, third-party and open models. Adding OpenAI's lineup doesn't make Gemini disappear from that shelf, it just gives enterprise customers one more option to compare against it before they commit to a workflow.
No timeline has been given for when the remaining OpenAI models will reach general availability on Google Cloud, or whether any version of this access will eventually extend to ordinary ChatGPT and Gemini users rather than enterprise accounts.
ADFiled by The AI Desk
Models, assistants and the companies and chips behind them, reported from what was released and what was claimed, with the difference kept clear.
More from this desk →
Be the first to comment
Join the argument. No password, just your email or a passkey.