Two of the biggest rivals in artificial intelligence just became business partners. On Wednesday, October 21, OpenAI and Google Cloud announced that OpenAI’s frontier models and its Codex agentic coding workflows are coming to Google Cloud’s Model Garden, the catalog where Google’s enterprise customers pick and deploy AI models.

It’s a notable shift for an industry where the major cloud platforms have tended to line up behind one favored model maker. Here’s what was announced, what’s available now and what is still on the way.

What was announced

According to OpenAI’s announcement, the partnership gives Google Cloud customers native access to OpenAI models for building “complex agentic workflows at scale,” backed by Google Cloud’s data governance and infrastructure. Customers will be able to buy the models through Google Cloud Marketplace, and companies with existing Google Cloud spending commitments can draw those commitments down on OpenAI usage.

Google’s own Google Cloud blog post lists three OpenAI models joining Model Garden:

  • GPT-6 Astra, described as designed for demanding work across coding, computer use, professional tasks and scientific research.
  • GPT-6.1 Sol, pitched as near-Astra intelligence at a lower cost for agentic coding, computer use and professional work.
  • GPT-6 Luna, a fast, cost-efficient model for everyday work at scale.

Google says these models are rolling out now in private preview to select enterprise customers, and that customers can also connect their local Codex CLI and Codex desktop workflows to Google’s Gemini platform for inference. OpenAI says general availability is planned “in the coming weeks.”

Coming soon: Antigravity and Gemini for business

The partnership goes beyond the model catalog. Both companies say OpenAI models will soon show up inside Google Antigravity, Google’s agent-focused development environment, and in Google’s Gemini business offerings. In practice, that would let a company standardized on Google tools choose an OpenAI model for a particular task without leaving the Google ecosystem.

Google Cloud CEO Thomas Kurian framed the deal around choice, saying the company wants to give customers “a broad selection of models, including those from OpenAI, so they can choose the right model for any given job.” OpenAI President Greg Brockman called it “an important step in making our most capable models and Codex available where customers are already building.”

Why it matters

The headline claim from Google is that it is now the only hyperscale cloud provider offering access to all three leading closed-source large language models, alongside more than 200 first-party, third-party and open models on one platform. For enterprise buyers, that pitch is about flexibility: swap models based on performance, speed and cost without signing separate vendor contracts or rebuilding an app around a single provider.

There are practical perks for finance teams too. Because OpenAI usage is billed through Google Cloud Marketplace, Google says it appears on a single Google Cloud invoice, counts against existing cloud commitments and applies toward its Flexible Savings Plans.

Security and compliance are a big part of the sales pitch. Google says customer prompts, context and outputs stay within the customer’s designated data perimeter and are never used to train foundation models. It also says OpenAI models run on dedicated Google Cloud infrastructure, with queries routed through isolated gateways and covered by unified audit logging. Some features, including Zero Data Retention options and Confidential Compute, are listed as coming soon rather than available today.

What it doesn’t change

The deal is not a breakup with Microsoft. OpenAI’s announcement states plainly that “Microsoft remains OpenAI’s primary cloud partner.” Instead, the move fits a broader pattern of AI companies making their models available wherever customers already work, rather than forcing businesses to move their data and tools to a single cloud.

For Google, it is a pragmatic bet. The company continues to develop its own Gemini models, yet it is clearly willing to host a direct competitor if that keeps enterprise workloads on Google Cloud.

What to watch next

For most people, nothing changes overnight. The rollout is a private preview aimed at select enterprise customers, and interested companies are being asked to fill out an interest form. The key things to watch over the coming weeks are when general availability arrives, how pricing compares with buying directly from OpenAI, and how quickly OpenAI models appear inside Antigravity and Gemini for business.

The bottom line

OpenAI’s arrival on Google Cloud is another sign that the AI race is becoming less about exclusive alliances and more about giving businesses choice. If you build on Google Cloud, you may soon be able to mix Gemini and GPT-6 models in the same workflow, on the same bill. If you don’t, the deal is still a useful signal of where enterprise AI is heading: toward a world where the model is a component you swap, not a platform you are locked into.

Image: Computer Servers by Open Grid Scheduler / Grid Engine via Flickr, CC0 1.0