Google Cloud adds OpenAI’s GPT-6 models to its model garden
Three OpenAI models enter private preview on Google’s AI platform, where enterprise customers get them under existing Google Cloud contracts and data controls rather than a separate OpenAI agreement
- Usage is transacted through Google Cloud Marketplace and appears on a single Google Cloud invoice
- Spending draws down existing cloud commitments and counts toward Flexible Savings Plans across model providers
- Zero data retention options and Confidential Compute are listed as coming soon
- OpenAI models are due to reach Gemini for business and Google Antigravity at a later date
- The blog gives no prices or regions
For a business already committed to Google Cloud spending, OpenAI’s models become something it can draw against an existing budget instead of a new supplier relationship with its own procurement, security review and contract. Moving between model families for different tasks becomes a configuration choice inside one platform, which weakens the case for signing directly with each model maker. Access is limited to selected enterprise customers, which leaves smaller firms without a way to try the models on Google’s terms, and without published prices the cost cannot be set against buying from OpenAI directly.
“Modern enterprise AI requires matching each workload to the right model rather than relying on a single model for every task.”Michael Gerstenhaber, vice-president of product management, Gemini, Google Cloud
Google Cloud began rolling out three OpenAI models – GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna – in private preview to select enterprise customers on 21 October, according to a company blog post by Michael Gerstenhaber, vice-president of product management, Gemini, at Google Cloud. The models sit in its model garden, and Google says more will follow.
Google describes Astra as built for demanding coding, computer use, professional and scientific work, Sol as close to Astra’s capability at lower cost, and Luna as a fast, cost-efficient model for everyday work at scale. From launch, customers can build with the models through Google’s APIs and route local Codex CLI and Desktop workflows to the Gemini platform for inference.
The company says the addition makes Google Cloud the only hyperscale cloud provider offering all three leading closed-source model families, alongside more than 200 first-party, third-party and open models. Google names those families as Gemini, Claude and GPT, as The Decoder also reported.
On data handling, Google says prompts, context and outputs stay inside the customer’s designated data perimeter and are not used to train foundation models, and that the OpenAI models run on dedicated Google Cloud infrastructure with queries routed through isolated gateways. Box, the content management company, is named as a customer. Yashodha Bhavnani, its vice-president of AI products, said the arrangement would pair OpenAI’s models with Google Cloud’s security and compliance capabilities.
Sources
- Google Cloud, “Google Cloud expands model choice and flexibility, bringing OpenAI models to customers”, blog post by Michael Gerstenhaber, 21 October 2026. https://cloud.google.com/blog/products/ai-machine-learning/google-cloud-brings-openai-models-to-customers
- The Decoder, “Google Cloud now offers OpenAI models alongside Gemini and Claude, becoming the only major provider with all three”, 21 October 2026. https://the-decoder.com/google-cloud-now-offers-openai-models-alongside-gemini-and-claude-becoming-the-only-major-provider-with-all-three/
