Anthropic made two announcements for business customers this week. On October 23 it merged chat and Cowork into a single experience for Claude Enterprise. A day earlier it made the full Claude Platform available on Google Cloud, operated by Anthropic itself.

One Claude for Enterprise

Enterprise users no longer pick a mode before they start. A conversation begins as a chat; when a task needs code execution or file handling, Claude continues in the cloud and returns the finished document, analysis or deck. Cloud tasks keep running after you close your laptop and can be resumed on web, desktop or mobile. Existing chats, projects, connectors, skills and plugins carry over. Pro and Max plans got the same merge in September.

Each cloud session runs in an isolated virtual machine with network access off by default for Enterprise, and connector credentials are kept outside the sandbox. Data is not used for model training by default.

Cost controls

Pricing stays usage-based. Anthropic says Opus 5.5 costs 40% less to run than Opus 5 on typical workloads, and Sonnet 5.5 up to 30% less than Sonnet 5 for most work; these are the company's own estimates. Because agent tasks use more tokens than a simple chat, admins get four controls: a new Efficient mode that keeps the unified experience on Sonnet and Haiku models, model access by role, spend limits at organization, group and member level, and pooled monthly budgets for groups. The update is generally available now. Customers using zero data retention for Claude Code, a HIPAA-ready configuration or customer-managed encryption keys stay on their current setup.

Claude Platform on Google Cloud, run by Anthropic

On October 22 Anthropic made Claude Models on Google Cloud (Anthropic managed) generally available. Google Cloud customers sign in with their existing IAM credentials, pay on one Google Cloud invoice and can count usage against existing Google Cloud commitments, while Anthropic operates the service and acts as data processor. New features and betas arrive the same day as on Anthropic's own platform, such as Managed Agents (beta), code execution, computer use and web search. Available models are Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5, on US and Global endpoints. Fast Mode for Opus 5.5 is generally available there, with up to 2.5 times faster output at $8 per million input tokens and $40 per million output tokens. The Google-managed Claude offering remains for teams that need Google as processor, strict regional data residency or integrations such as Model Armor.

What this means if you buy Claude for a team

If your company is on Claude Enterprise, set spend limits and decide who gets Opus before the rollout reaches everyone. Efficient mode is the simplest way to cap costs for staff who mostly write and summarize.

If your developers already pay for Google Cloud, the Anthropic-managed option gives them new Claude features on day one on the bill they already have. For coding work specifically, see our Claude Code card.