Claude's Full Platform Reaches Google Cloud With Day-Zero Features
Google Cloud customers can now buy the full Claude Platform through an Anthropic-managed offering, with new features and betas arriving the same day they launch natively. Anthropic announced general availability of Claude Models on Google Cloud on October 22. Customers sign in with existing Google Cloud IAM credentials and receive a single Google Cloud invoice that draws down existing commitments.
The feature list is the full platform: Claude Managed Agents (beta), code execution, computer use, agent skills, an advisor strategy (beta), the Files API, citations, web search and fetch, an MCP connector (beta), prompt caching, batch processing and the Claude Console. Fast Mode on Claude Opus 5.5 is now generally available, with up to 2.5x faster output at 8 dollars input and 40 dollars output per million tokens. The models offered are Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5. Triple Whale’s Maksim Gorev said prompt caching is central to running its Moby harness cost-effectively.
The decision that matters is which of the two deployment options to pick. Anthropic managed gives day-zero feature access, with Anthropic as data processor. Google managed fits strict regional data residency or native integrations such as Model Armor. That is a trade between speed and control, and it is not new: when Anthropic brought the full Claude Desktop experience to AWS, Google Cloud and Microsoft Foundry in June, inference stayed inside the customer’s own cloud environment and region.
For a company already committed to Google Cloud, the procurement barrier to the latest Claude capabilities is gone. Pick per workload: regulated data goes on the Google-managed path, and teams piloting agents that need beta features go Anthropic-managed. Make that call before the pilot starts, because it determines who processes your data.