# Claude and OpenAI models through Google Cloud: what changes

URL: https://infro.io/blog/claude-openai-models-through-google-cloud
Author: INFRO Editorial (https://infro.io/blog/editorial)
Publisher: INFRO
Published: 2026-10-27
Last modified: 2026-10-27
Last reviewed: 2026-10-27 (next review due 2026-11-26)
Topic: Architecture and migration
Format: Comparison

> Claude Models on Google Cloud (Anthropic managed), launched 22 October 2026, is the Claude API operated by Anthropic and reached through your Google Cloud project: Google Cloud IAM controls access, Google Cloud Marketplace bills it, usage counts toward a Google Cloud spend commitment like other Marketplace purchases, and it supports Files, batches, server tools and Managed Agents. Claude on Vertex AI, now called Claude Models on Google Cloud (Google managed), is operated by Google with Google's quotas and a smaller feature set. OpenAI's models and Codex are in private preview in Model Garden, with general availability planned in the coming weeks and commitment drawdown through Marketplace.

Key takeaways:
- Claude Models on Google Cloud (Anthropic managed) is the Claude API run by Anthropic at claude.googleapis.com; Google Cloud handles access and billing, and Anthropic is the data processor.
- Claude on Vertex AI is now called Claude Models on Google Cloud (Google managed); Google operates it, sets its quotas and lifecycle dates, and it lacks the Files API, Message Batches, code execution, web fetch and Managed Agents.
- Anthropic-managed usage bills through Google Cloud Marketplace in Claude Consumption Units of $0.01 and counts toward a Google Cloud spend commitment the way other Marketplace purchases do.
- Setting up the Anthropic-managed offering creates a new Anthropic organization: API keys, workspaces, files, batches and Zero Data Retention arrangements do not carry over, and it accepts no API keys at all.
- OpenAI's GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna are in private preview in Model Garden for select enterprise customers, with general availability planned in the coming weeks; Zero Data Retention and Confidential Compute are listed as coming soon.

On 22 October 2026 Anthropic launched Claude Models on Google Cloud (Anthropic managed): in Anthropic's words, "the Claude API operated by Anthropic and reached through your Google Cloud project", with Google Cloud IAM controlling access and Google Cloud Marketplace handling billing[1]. It is not Claude on Vertex AI. That offering, which Anthropic's documentation now calls Claude Models on Google Cloud (Google managed), is run by Google as part of its Agent Platform[2]. A day earlier, OpenAI and Google announced OpenAI's frontier models and Codex for Google Cloud's Model Garden, in private preview, transacted through Google Cloud Marketplace and able to draw down Google Cloud commitments[3][4].

For a developer the choice changes the endpoint, the authentication, the feature set and the quotas. For a CFO it changes whose invoice the spend appears on, which commitment it burns down, and who processes the data. This comparison covers both, and the cases where each option loses. It extends [Claude on the Anthropic API, Bedrock or Google Cloud](https://infro.io/blog/claude-bedrock-vertex-or-anthropic-api), written before the Anthropic-managed offering existed.

## The options side by side

|  | Claude API (direct) | Claude on Google Cloud, Anthropic managed | Claude on Google Cloud, Google managed (Vertex AI) | Claude in Amazon Bedrock | OpenAI models in Model Garden |
| --- | --- | --- | --- | --- | --- |
| Who operates inference | Anthropic | Anthropic | Google | AWS | Not stated; Google says they run on dedicated Google Cloud infrastructure |
| Data processor | Anthropic | Anthropic | Google Cloud governs data handling | AWS | Not stated |
| Endpoint and format | api.anthropic.com, Claude API | claude.googleapis.com, Claude API paths | Agent Platform `:rawPredict` | Bedrock runtime | Model Garden |
| Authentication | Anthropic API keys | Google Cloud IAM only, no API keys | Google Cloud IAM | AWS IAM or Bedrock bearer tokens | Not stated |
| Rate limits set by | Anthropic | Anthropic | Google Cloud quotas | AWS | Not stated |
| Billed through | Anthropic | Google Cloud Marketplace | Google Cloud, as Agent Platform usage | AWS | Google Cloud Marketplace |
| Status | Generally available | Launched 22 Oct 2026; SDK client in beta | Generally available | Generally available | Private preview |

Table: Ways to reach Claude and OpenAI models through a cloud account, from Anthropic's, Google Cloud's and OpenAI's pages read on 2026-10-27.

The cells come from Anthropic's pages on the two Google Cloud offerings[5][6][2], its Bedrock and Claude Platform on AWS pages[7][8], and the two announcements about OpenAI models[4][3]. Anthropic also offers an AWS counterpart of the new offering: Claude Platform on AWS, which Anthropic operates, with AWS handling authentication, IAM access control and billing through AWS Marketplace[8].

## Who runs the model and who sees the data

This is the row a security review asks about first.

**Anthropic managed.** "Anthropic is the data processor" for this offering, and Anthropic notes that "some supporting services, such as Anthropic's trust and safety systems, process data outside Google Cloud"[5]. Data retention follows the Claude API's policy; Zero Data Retention is available on request, and a ZDR arrangement on another Anthropic organization "doesn't carry over"[5]. Google Cloud's gateway at `claude.googleapis.com` checks your credentials and IAM permission, then forwards the request to Anthropic[5].

**Google managed.** "Google Cloud governs data handling" for Claude on Agent Platform, and prompt and completion logging is Agent Platform's own request-response logging service[6]. If your requirement is that the cloud provider is the only party processing inference, this is the Google Cloud option that fits it, as Bedrock is on AWS: Anthropic describes Bedrock as running "entirely on AWS-controlled infrastructure with AWS as the operating party"[8].

**OpenAI on Google Cloud.** Google's announcement says OpenAI models "run on dedicated Google Cloud infrastructure", that queries are routed through isolated gateways within Google Cloud, and that prompts, context and outputs "remain entirely within your designated data perimeter and are never used to train foundation models"[4]. It also lists Zero Data Retention options and Confidential Compute as "coming soon"[4]. Neither announcement says which company operates the serving stack, so ask before you file the processing record.

## Features: where the Google-managed and Bedrock versions lose

The Anthropic-managed offering lists, as working the way they do on the Claude API: fast mode for Claude Opus 5.5, tool use including web search, web fetch, code execution, the memory tool, the advisor tool and the MCP connector, Agent Skills and the Files API, prompt caching, extended thinking, server-side fallback, structured outputs, token counting, the Message Batches API and the Models API, plus Claude Managed Agents[5]. What it does not support: API keys, MCP tunnels, memory stores on self-hosted environments, and the pre-built self-hosted sandbox workers in the SDKs and CLI[5]. Context windows match the Claude API[5].

The Google-managed offering does not support URL sources for images and documents, the Files API, code execution, web fetch, the advisor tool, Agent Skills, the MCP connector, programmatic tool calling, the Message Batches, Models, Admin, Compliance and Usage and Cost APIs, Claude Managed Agents, or server-side fallback[6]. Anthropic's Bedrock page lists a near-identical gap, adding web search, automatic prompt caching and the newest computer-use and browser-use toolsets[7].

The practical reading: if your code uses batches, files, server tools or Managed Agents, the Google-managed and Bedrock routes need a rewrite or a second path, and the Anthropic-managed route does not.

## Quotas, locations and residency pricing

**Rate limits.** On the Anthropic-managed offering, "Anthropic manages rate limits"; higher limits come from your Anthropic account representative or sales[5]. On the Google-managed offering, Google Cloud manages quotas and quota increases[6]. Check limits before moving production traffic; Anthropic's migration steps say the same[5].

**Location.** The Anthropic-managed offering uses the `global` location, which is separate from where inference runs; for US-only inference you set `inference_geo: "us"` per request[5], which carries a 1.1x multiplier on all token prices for Claude 4.6 and later models on the Claude API, Claude Platform on AWS and this offering[9]. The Google-managed offering has global, multi-region (`us`, `eu`) and regional endpoints; multi-region and regional endpoints cost 10 percent more than global, and regional endpoints serve Claude Sonnet 4.6 and earlier only[6]. Bedrock's regional endpoints also carry a 10 percent premium[7].

**Lifecycle.** Google sets the retirement dates for models on the Google-managed offering, and they can differ from the Claude API's; Anthropic's own dates apply to the Anthropic-operated platforms[6].

## Billing and commitments: the CFO's view

The reason most enterprises will look at these offerings is the invoice, so the details matter:

- **Anthropic managed.** It "bills through Google Cloud Marketplace". Anthropic meters usage and reports it to Google Cloud in Claude Consumption Units, where one unit is $0.01; charges appear on your Google Cloud bill and "Anthropic doesn't send a separate invoice". Usage and cost are visible in the Claude Console. And: "If you have a Google Cloud spend commitment, usage counts toward it the way other Marketplace purchases do"[5].
- **Google managed.** Google Cloud invoices it as Agent Platform usage, at Google Cloud's published prices, which depend on the endpoint type[6].
- **OpenAI on Google Cloud.** OpenAI says enterprises with Google Cloud spend commitments "will be able to draw down their GCP commitments for these purchases"[3]. Google adds that usage counts toward Flexible Savings Plans across model providers[4].

Two cautions. First, setup of the Anthropic-managed offering creates a new Anthropic organization, separate from any you have; Anthropic asks customers with existing negotiated pricing to contact their Anthropic or Google Cloud representative before starting, because those terms "can't be applied to past usage"[5]. Do that before the first request, not after the first bill. Second, a commitment drawdown is not a lower rate: it moves spend you already promised to Google onto AI usage. Whether that is worth it depends on how much of the commitment you would otherwise leave unused.

## Moving a workload: what the code change is

From the direct Claude API, three things change and the request body does not[5]:

- **Base URL:** `https://claude.googleapis.com/v1/projects/{project}/locations/global/workspaces/{workspace}/invoke`, then the usual path such as `/v1/messages`.
- **Authentication:** an `Authorization: Bearer` header with a Google access token; no `x-api-key`.
- **Workspace:** in the URL, not in an `anthropic-workspace-id` header.

```python
# pip install -U "anthropic[google_cloud]"
# Env: ANTHROPIC_GOOGLE_CLOUD_PROJECT, ANTHROPIC_GOOGLE_CLOUD_WORKSPACE_ID (starts wrkspc_)
# The caller needs roles/claude.user on the project; credentials come from ADC.
from anthropic import AnthropicGoogleCloud

client = AnthropicGoogleCloud()
message = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello, Claude"}],
)
print(message.content)
```

The SDK client for this offering is in beta, and it ignores `ANTHROPIC_API_KEY`, `ANTHROPIC_AUTH_TOKEN` and `ANTHROPIC_BASE_URL`[5]. Before cut-over: enable `claude.googleapis.com` and the Marketplace service on every calling project, grant `roles/claude.user` to each calling identity, recreate workspaces, and remember that files, skills, batches and Managed Agents resources stay in the old organization[5]. Turn on Data Access audit logs for `claude.googleapis.com` if you want a record of requests; they are off by default[5].

From Vertex AI, the change is the client class (`AnthropicVertex` to `AnthropicGoogleCloud`), the model id format for models before Claude 4.6 (`claude-haiku-4-5@20251001` becomes `claude-haiku-4-5-20251001`), and a check of the endpoint location you rely on today[5].

## How to choose

| If this matters most | Choose | It loses when |
| --- | --- | --- |
| Every Claude feature, Anthropic's quotas, no cloud dependency | Claude API direct | You need the spend on your Google Cloud or AWS bill, or Google Cloud IAM for access |
| Full Claude API features, billed and access-controlled through Google Cloud | Anthropic managed | You need Google to be the only processor, an API key based client, or MCP tunnels |
| Google as the operating party, regional endpoints, Google quotas | Google managed | Your code uses batches, files, code execution, web fetch or Managed Agents |
| AWS as the operating party | Bedrock | Same feature gaps as Google managed |
| OpenAI models on a Google Cloud commitment | OpenAI in Model Garden | You need it in production now: it is in private preview, with ZDR listed as coming soon |

Table: Choosing a route for Claude or OpenAI models, by what decides the case.

For OpenAI models outside Google Cloud, the other cloud route is Azure; OpenAI's announcement states that "Microsoft remains OpenAI's primary cloud partner"[3]. The trade-offs there are in [Azure OpenAI vs the OpenAI API](https://infro.io/blog/azure-openai-vs-openai-api), and running several clouds behind one interface is covered in [one gateway across Azure, Bedrock and more](https://infro.io/blog/multi-cloud-azure-bedrock-one-gateway).

## Which of these INFRO connects

INFRO is a control plane that sits between your applications and provider accounts you already hold; the provider or the cloud bills you directly, and INFRO charges a separate subscription. Today INFRO connects Claude through your own Anthropic account and through Google Vertex AI, that is the Google-managed offering: you give INFRO a service-account key for your project, it signs short-lived access tokens with it, and the account needs only `roles/aiplatform.user`. Your application names `anthropic/claude-opus-5.5`, and a [route](https://infro.io/docs/routes) can order or weight the two connections, so a failure on one moves the request to the other before the response starts.

What INFRO does not connect yet, stated plainly: the Anthropic-managed Google Cloud offering at `claude.googleapis.com`, Claude in Amazon Bedrock, and OpenAI models in Model Garden. If you move Claude traffic to the Anthropic-managed offering, calls made there do not pass through INFRO and do not appear in its usage or budgets. INFRO also does not change which commitment your spend draws down: that is decided by the account the provider or cloud bills.

## Questions

### Is Claude Models on Google Cloud (Anthropic managed) the same as Claude on Vertex AI?

No. The Anthropic-managed offering is the Claude API run by Anthropic, reached at claude.googleapis.com with the Claude API's request format. Claude on Vertex AI, which Anthropic now calls Claude Models on Google Cloud (Google managed), is run by Google on its Agent Platform prediction endpoints, with Google's quotas, lifecycle dates and a smaller feature set. Both bill through Google Cloud and use Google Cloud credentials.

### Does Claude usage through Google Cloud count toward our Google Cloud commitment?

For the Anthropic-managed offering, Anthropic's documentation says usage counts toward a Google Cloud spend commitment the way other Marketplace purchases do. Confirm with your Google Cloud account team how your specific commitment treats Marketplace purchases before you move spend.

### Can I use an API key with the Anthropic-managed Google Cloud offering?

No. It authenticates with Google OAuth access tokens from Application Default Credentials, service accounts or Workload Identity Federation, and access is granted with the roles/claude.user, roles/claude.admin and roles/claude.viewer IAM roles. The Claude Console cannot create API keys for the organization it sets up.

### Can I buy OpenAI models through Google Cloud today?

Only in private preview. Google says GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna are rolling out to select enterprise customers in Model Garden, and OpenAI says general availability is planned in the coming weeks. Interested customers register through Google's interest form.

### Do my existing Anthropic files, batches and keys move to the Google Cloud offering?

No. Setup creates a new Anthropic organization for the Google Cloud project. API keys, workspaces and console settings do not carry over, files, skills, batches and Managed Agents resources stay in the existing organization, and a Zero Data Retention arrangement on another organization does not apply. The two organizations can run in parallel while you move traffic.

## Sources

1. Claude Platform release notes — Anthropic. https://platform.claude.com/docs/en/release-notes/overview (accessed 2026-10-27)
2. Claude Models on Google Cloud — Anthropic. https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/overview (accessed 2026-10-27)
3. OpenAI models are coming to Google Cloud — OpenAI. https://openai.com/index/openai-models-google-cloud/ (accessed 2026-10-27)
4. Google Cloud expands model choice and flexibility, bringing OpenAI models to customers — Google Cloud. https://cloud.google.com/blog/products/ai-machine-learning/google-cloud-brings-openai-models-to-customers (accessed 2026-10-27)
5. Claude Models on Google Cloud (Anthropic managed) — Anthropic. https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/anthropic-managed (accessed 2026-10-27)
6. Claude Models on Google Cloud (Google managed) — Anthropic. https://platform.claude.com/docs/en/build-with-claude/claude-models-on-google-cloud/google-managed (accessed 2026-10-27)
7. Claude on Amazon Bedrock — Anthropic. https://platform.claude.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy (accessed 2026-10-27)
8. Claude Platform on AWS — Anthropic. https://platform.claude.com/docs/en/build-with-claude/claude-platform-on-aws (accessed 2026-10-27)
9. Pricing — Anthropic. https://platform.claude.com/docs/en/about-claude/pricing (accessed 2026-10-27)