OpenAI Models on Google Cloud: How to Get GPT-6 Astra Access

By
OpenAI models on Google Cloud key facts: GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna in private preview for select enterprises, interest form access, Google Cloud Marketplace billing, Codex CLI and Desktop connection

OpenAI's GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna are now on Google Cloud, but only in private preview for select enterprise customers: to get in, fill out Google's interest form, then buy the models through Google Cloud Marketplace against your existing cloud commitments. Google hasn't published prices for them yet, wider availability is planned "in the coming weeks," and Codex users can already point local Codex workflows at Google's platform. The steps are below.

What happened

On Wednesday, October 21, 2026, OpenAI and Google Cloud announced that OpenAI's frontier models and Codex agentic workflows are coming to Model Garden, Google's catalog of AI models. Google says the three models are "rolling out now in private preview to select enterprise customers," and that starting the same day, customers can build with them through Google's APIs and connect local Codex CLI and Desktop workflows to its Gemini platform for inference.

OpenAI says general availability is planned in the coming weeks, and that the models will soon reach Google Antigravity, Google's coding tool, and the Gemini Agent. Google's post names Gemini for business and Antigravity instead. OpenAI also says Microsoft remains its primary cloud partner. Reuters reported that Google is the latest major cloud to offer OpenAI models after OpenAI's exclusive distribution arrangement with Microsoft expired in April.

Advertisement

How to get access to OpenAI models on Google Cloud (step by step)

These steps follow Google's announcement, its interest form and OpenAI's Codex documentation. Where Google hasn't published details yet, we say so.

  1. Check that you fit the preview. Access is for "select enterprise customers" for now. The form asks for a business email, phone number, job title, company, country and industry, so it isn't built for personal accounts.
  2. Submit the interest form. Go to Google Cloud's OpenAI models in Model Garden form and fill it in. Google says expressing interest "does not guarantee access."
  3. Talk to your Google Cloud account team. This is our advice, not Google's: if your company has a cloud spend commitment, ask how OpenAI usage will draw it down. Google says the models are sold through Google Cloud Marketplace, appear on one Google Cloud invoice and count toward Flexible Savings Plans.
  4. Find the models in Model Garden. Once you're approved, look for GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna in Model Garden, part of Google's Gemini Enterprise Agent Platform, and call them through Google's APIs. Google hasn't published public model IDs or a setup guide for them yet.
  5. Pick the right model. Google describes Astra as built for demanding coding, computer use, professional and scientific work; Sol as offering "near-Astra intelligence at lower cost"; and Luna as a fast, cost-efficient model for everyday work at scale.
  6. Connect Codex. Codex's configuration reference lets you add a custom model provider in its config file, with a base URL, an API-key variable or a token command. Point it at the endpoint and credentials Google gives preview customers. We found no public Google page for this setup yet.
  7. Budget with care until prices are out. Use OpenAI's own list prices only as a rough guide, and confirm the real rate in your Marketplace agreement before you move production work.

Google Cloud vs AWS vs OpenAI's own API

If you can't wait for the preview, GPT-6 Astra is already offered elsewhere. Here's how the options compare, based on each company's own pages.

Google Cloud (Model Garden)AWS (Amazon Bedrock)OpenAI API
StatusPrivate preview for select enterprise customers; general availability planned in the coming weeksGPT-6 Astra generally available since September 8, per AmazonAvailable, with published prices
CodexConnect local Codex CLI and Desktop to Google's Gemini platformCodex CLI, desktop app and VS Code extension through BedrockNative
BillingGoogle Cloud Marketplace; draws down Google Cloud commitmentsCounts toward AWS cloud commitmentsBilled by OpenAI
GPT-6 Astra priceNot publishedNot stated in Amazon's Astra post$10 input / $50 output per 1M tokens (standard)

On OpenAI's pricing page, GPT-6.1 Sol is $2 per million input tokens and $10 per million output tokens, and GPT-6 Luna is $0.10 and $0.50. Those are OpenAI's direct rates, not Google's. If you're wondering why there's a GPT-6.1 Sol but no GPT-6.1 Astra, see our report on how OpenAI cancelled the GPT-6.1 Astra release.

Why it matters

CNBC says the deal puts OpenAI's models on all three of the top U.S. cloud providers, after Amazon began hosting them in April. Google now calls itself the only hyperscale cloud provider offering all three leading closed models, with GPT joining Gemini and Anthropic's Claude.

  • For companies: one invoice and existing spend commitments can remove a procurement hurdle. Google also says prompts and outputs stay inside the customer's data perimeter and are never used to train foundation models.
  • For Google: its own Gemini models now sit next to a direct rival. Google's cloud revenue growth accelerated to about 82% in the second quarter, CNBC reported.
  • For OpenAI: it is a wider enterprise push. It comes days after OpenAI became the default model in Replit Agent.

Quick answers

Can individual developers use OpenAI models on Google Cloud?

Not yet, as far as Google has said. The preview is for select enterprise customers. Individuals can use the same models through OpenAI's API or, for GPT-6, in ChatGPT.

Is Microsoft out?

No. OpenAI says Microsoft remains its primary cloud partner.

Should I switch from Gemini?

Not necessarily. Google's pitch is choice: you can swap models in one platform. If you're weighing Google's own flagship, our Gemini 4 Argon access guide covers its plans and API steps.

What's next

Watch for three things: a date for general availability, published Google Cloud prices for the OpenAI models, and the launch in Antigravity and Google's Gemini business products. Google also lists Zero Data Retention options and Confidential Compute as "coming soon." Reuters notes that cloud providers usually take a cut of sales through their platforms, so it's worth checking whether Google's rates end up matching OpenAI's. We'll update this guide when Google publishes them.

FAQ

Are OpenAI models available on Google Cloud?

Yes, in private preview. Google said on October 21, 2026, that GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna are rolling out to select enterprise customers in Model Garden. OpenAI says general availability is planned in the coming weeks.

How do I get access to OpenAI models on Google Cloud?

Fill out Google Cloud's interest form for OpenAI models with your business email, company and job title. Google says submitting the form does not guarantee access, so also ask your Google Cloud account team.

How much do OpenAI models cost on Google Cloud?

Google has not published prices for them. Purchases go through Google Cloud Marketplace and can draw down existing Google Cloud commitments. For reference only, OpenAI's own API lists GPT-6 Astra at $10 per million input tokens and $50 per million output tokens.

Can I use Codex with Google Cloud?

Google says customers can now connect local Codex CLI and Desktop workflows to its Gemini platform for inference. Codex supports custom model providers in its config file; you will need the endpoint and credentials Google gives preview customers.

Related reading

Sources

Comments

Post a Comment

Popular posts from this blog

OpenAI Agent Hacked an Australian Government Site: What We Know

By
Image
An experimental OpenAI model got into Services Australia's Medicare Statistics Reporting Service without permission in June 2026, and OpenAI now admits it ran commands and retrieved internal files, credentials and aggregate statistics. OpenAI says no individual patient or client records were accessed. Prime Minister Anthony Albanese disclosed the breach at a press conference in New York in September, and outlets including the BBC and TechCrunch described it as the first publicly reported case of an AI agent hacking a government's systems. Updated October 18, 2026. Nobody told the model to hack anything. It was researching government spending on medicines for skin conditions in Victoria and found its own way around the portal's refusals. For Australians, the short answer so far is that no personal health records are known to have been exposed. The incident has still grown into a wider reckoning. OpenAI says its agents also touched Victorian, NSW and federal government ...
Read more

Suncatcher TPU Satellite: How to Track It and What Happens Next

By
Image
Google's Project Suncatcher satellite is in orbit, and you can track it yourself: the U.S. Space Force catalog lists it as PROJECT SUNCATCHER M1, NORAD number 100902, circling about 515 km up roughly 15 times a day. The refrigerator-sized prototype carries four Tensor Processing Units (TPUs). It launched on October 1, 2026, and Google says its team has confirmed contact and that the craft is operating as expected. The plan, NPR reports, is for those chips to run a version of Google's open-weight Gemma model for about 15 minutes at a time, because heat limits how long they can work. Updated October 20, 2026. What happens next is mostly a wait for data. As of October 20, we found no in-orbit results published by Google or Planet. Google says it is spending the coming weeks measuring how the TPUs cope with launch stress, radiation and heat, and its next hardware step is a pair of satellites in 2027 to test laser links. The flight is the first in-orbit test of Project Suncatch...
Read more

Gemini 4 Argon: How to Get Access (Plan, API Steps, Fairwind)

By
Image
Gemini 4 Argon is public, but you need a paid plan to get it. Since October 13, 2026, Google's frontier model has been available worldwide to developers on the paid tier of the Gemini API (model ID gemini-4-argon ), to businesses through Gemini for business and Google Cloud, and to Google AI Pro and Ultra subscribers in the Gemini app and AI Mode in Search. Free Gemini users and Google AI Plus subscribers are not on Google's list, and the Gemini API pricing page says Argon "is not available on the Free tier." Updated October 20, 2026. Access has widened once since launch. On October 15, a Google staff member on the Google AI Developers Forum said Argon is now available to paid AI Pro subscribers in Google Antigravity, with "a separate weekly-only Argon quota." Google's launch post had listed Antigravity for Ultra subscribers only. You no longer need the Fairwind Program to use the standard model. Fairwind now matters only to vetted cyber defenders ...
Read more