Google Cloud adds OpenAI models to its model garden in private preview

Three OpenAI models, three workload profiles

Google Cloud says the launch is meant to let enterprises match each workload to a suitable model instead of relying on one model for every task. Michael Gerstenhaber, VP of Product Management for Gemini at Google Cloud, wrote that the company is "expanding that foundation by launching OpenAI models in Google Cloud’s model garden."

The three models are described as follows:

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

Google Cloud claims it is now the only hyperscale cloud provider offering access to all three leading closed source LLMs, alongside more than 200 first-party, third-party and open models on a unified enterprise AI platform.

Billing, switching and governance

According to Google Cloud, customers can dynamically select and swap models for any use case based on performance, latency and cost efficiency, without architectural lock-in or separate vendor contracts. Because OpenAI models are transacted directly through Google Cloud Marketplace, all usage appears on a single Google Cloud invoice, draws down existing cloud commitments and counts toward Flexible Savings Plans (FSP) across model providers.

The company also stresses security and compliance. Customer prompts, context and outputs are said to remain within the designated data perimeter and are never used to train foundation models. Zero Data Retention (ZDR) options are listed as coming soon.

OpenAI models run on dedicated Google Cloud infrastructure, with Confidential Compute also coming soon. Queries are routed through hardened, isolated gateways, and Google Cloud says third-party model providers have no unauthorized access to a customer's cloud environment. Every agent and model interaction operates under an authenticated identity, giving IT administrators visibility, VPC Service Controls protection and unified audit logging.

Availability and what comes next

The three models are in private preview for select enterprise customers, with more coming soon. Starting today, customers can build with them through Google Cloud's APIs and connect their local Codex CLI and Desktop workflows to the Gemini platform for inference. Interested customers are asked to use a form to express interest. The source gives no pricing or general availability date.

Google Cloud also says OpenAI models will come soon to Gemini for business and Google Antigravity, so that enterprise customers and developers can pick the best model for each task.

For enterprise buyers, the main practical change is procurement and operations: OpenAI models become available inside an existing Google Cloud setup, billed on the same invoice and covered by the same identity and audit controls. Access is limited for now to the private preview.