OpenAI's frontier models and Codex are coming to Google Cloud's Model Garden, with general availability planned in the coming weeks, according to a joint announcement on October 21, 2026. Customers will buy them through the Google Cloud Marketplace and can apply existing Google Cloud commitments to the spend. Neither company has disclosed pricing.
What happened, and when?
On October 21, 2026, OpenAI announced that its frontier models and Codex agentic workflows will reach Google Cloud. In OpenAI's announcement, the company says general availability is "planned in the coming weeks" and that OpenAI models will "soon be available in Google Antigravity and the Gemini Agent." Axios reported the deal the same day, describing purchases through Google's Model Garden and Marketplace, including with existing cloud spending commitments.
OpenAI's page spells out the money path: "Enterprises with Google Cloud spend commitments will be able to draw down their GCP commitments for these purchases." Neither source states pricing or regions. Treat any per-token figure you see in a forum thread as a guess until Google publishes a price sheet.
What the executives said
Greg Brockman, OpenAI's president and co-founder, called the partnership "an important step in making our most capable models and Codex available where customers are already building." Thomas Kurian, Google Cloud's chief executive, put the emphasis on choice: "We are focused on giving customers a broad selection of models, including those from OpenAI, so they can choose the right model for any given job, helping to deliver optimal quality and lower costs."
Axios also quoted Mike Leone, a vice president and analyst at Moor Insights & Strategy, who read the move as a sign of confidence from Google: "Gemini now has to win those customers with OpenAI sitting right next to it."
Why the Microsoft line matters
OpenAI's page includes a sentence worth a second read: "Microsoft remains OpenAI's primary cloud partner." Axios adds that changes to the Microsoft partnership opened the door to broader cloud distribution, and points to a separate OpenAI expansion on Amazon. Neither source gives exclusivity terms, dates or contract details. What the sources do show is a company that now sells its models through more than one hyperscaler marketplace.
What it means for solo operators and marketers
Most of you will never negotiate a cloud commitment. Your clients might. The practical shift is procurement. An enterprise client that already pays Google Cloud may be offered OpenAI models through a channel it already buys from, which skips a vendor-onboarding step. For agencies, that means two conversations: which model a workflow should use, and which budget line pays for it. Until pricing is published, do not quote a client a per-run cost.
Codex on Google Cloud also matters if you run agentic coding or automation for clients. Teams building on Google infrastructure can keep an agent next to their data instead of wiring in a separate vendor. A solo operator without a developer on staff should read this as a future option, not a Monday task.
Our take
I co-founded Urchin, the software that became Google Analytics, and I have spent a long time watching Google decide where marketers spend money. My read: the model is becoming a commodity, and the fight moves to the marketplace. When OpenAI, Anthropic and Gemini all sit on the same shelf, the buyer stops asking which model is smartest and starts asking which invoice they already pay. Mike Leone's point about data, governance and cost is the right frame, and it is the one I would put in front of any client.
Google gets the obvious win. Offering OpenAI's models inside a storefront it already runs is a clever way to keep enterprise customers from drifting. The number to watch is pricing. If Google matches OpenAI's direct price, the marketplace becomes the default. If it adds a margin, the savings story thins out. Until the price sheet lands, run the numbers on a real workload, not the keynote.