At present, we’re excited to announce Claude Cowork in Amazon Bedrock. Now you can run Cowork and Claude Code Desktop by Amazon Bedrock, straight or utilizing an LLM gateway.
From startups to international enterprises throughout each business, organizations construct with Claude Code in Amazon Bedrock to spice up developer productiveness and speed up supply. With Amazon Bedrock you’ll be able to construct inside your current AWS surroundings, preserve enterprise safety and regional information residency, and scale inference. Your information stays underneath your account’s controls: Amazon Bedrock doesn’t retailer prompts, information, instrument inputs and outputs, or mannequin responses, and doesn’t use them to coach basis fashions.
With Claude Cowork in Amazon Bedrock, you’ll be able to develop AI adoption to each information employee in your group, with a desktop software that reads paperwork, runs multi-step analysis, processes information, and returns completed work.
On this publish, we stroll by how Claude Cowork integrates with Amazon Bedrock and present an instance of how information employees use it in observe.
What’s Claude Cowork
With Claude Cowork, your customers can delegate analysis, doc evaluation, information processing, and report era to Claude from a desktop software. They get the core Claude Desktop capabilities together with initiatives, artifacts, reminiscence, file add and export, distant connectors, abilities, plugins, and MCP servers. Options that require Anthropic-hosted inference, together with the Chat tab, Pc Use, and the Abilities Market, should not included as a result of Claude Cowork routes mannequin inference solely by Amazon Bedrock in your AWS account. For a full function comparability with Claude Enterprise, see Options on 3P.
Pricing is consumption-based by your current AWS settlement and billing, with no seat licensing from Anthropic.
How Claude Cowork Integrates with Amazon Bedrock
Amazon Bedrock serves because the inference backend in your AWS account and supported AWS Areas.
Configuring Claude Cowork in Amazon Bedrock takes two steps. First, customers obtain the Claude Desktop software on their machine. Second, your device-management system (akin to Jamf, Microsoft Intune, or Group Coverage) pushes a configuration to Claude Desktop that prompts the inference mode, specifying the mannequin ID and Amazon Bedrock Inference Profile, authentication methodology, and organizational insurance policies. In case your group centralizes mannequin entry by an LLM gateway, you level Claude Desktop on the gateway URL by the identical managed configuration.
In case your group already builds with Claude Code in Amazon Bedrock, Claude Cowork can use the identical setup.
Determine 1: The next diagram illustrates the end-to-end circulation
The applying has three outbound paths, all underneath your management. Mannequin inference goes to Amazon Bedrock within the AWS Areas you configure. MCP server connections, if configured, go to endpoints you approve. Anthropic receives solely combination telemetry (token counts, mannequin ID, error codes, nameless system identifier), which might be disabled by configuration choices.
Amazon Bedrock provides in-Area, geo cross-Area, and international cross-Area inference profiles so you’ll be able to select the correct stage of knowledge residency in your group.
Claude Cowork works with the AWS providers you already use:
For particulars on MDM configuration, credentials, MCP servers, and plugins, see the Claude Cowork Configuration Reference.
Claude Cowork in observe
With the combination configured, your customers open Claude Desktop and begin delegating work. Claude Cowork can connect with exterior information sources by MCP servers, giving Claude entry to stay documentation, internet search, and different instruments as it really works.
For instance, a product supervisor is planning a brand new notification function for a college athletics app hosted on AWS. They’ve buyer assembly notes that time in several instructions, a set of challenge necessities, and restricted time to reconcile them. They add them to Cowork.
Claude compares the disparate inputs and synthesizes them right into a single product temporary. It evaluates the proposed strategy, analysis alternate options, flags technical challenges, and backs suggestions with proof factors. Related to the AWS Documentation MCP server and an internet search MCP server, Claude grounds the temporary in present service documentation, market context, and competitor positioning.
Determine 2: The product supervisor turns assembly notes right into a product temporary with Claude Cowork
In minutes, the product supervisor has a structured temporary grounded in present sources and prepared for evaluation. The identical sample applies to different information employees. An operations supervisor can consolidate scattered documentation into an SOP. A finance analyst can flip uncooked information right into a formatted month-to-month evaluation. A analysis group can compile findings from a number of sources right into a single report.
Conclusion
With Claude Cowork in Amazon Bedrock, you’ll be able to develop AI adoption to each information employee in your group whereas conserving your information inside your AWS surroundings. Claude Cowork is on the market on macOS and Home windows in AWS Areas the place Claude fashions can be found on Amazon Bedrock. To get began, obtain Claude Desktop from claude.com/obtain and see the Claude Cowork Setup Information.
Give Claude Cowork a strive at the moment and ship suggestions to AWS re:Put up for Amazon Bedrock or by your ordinary AWS Assist contacts.
Concerning the authors
Sofian Hamiti
Sofian Hamiti is a know-how chief with over 12 years of expertise constructing AI options, and main high-performing groups to maximise buyer outcomes. He’s enthusiastic about empowering numerous skills to drive international influence and obtain their profession aspirations.
Ayan Ray
Ayan Ray is a Principal Associate Options Architect and AI Tech Lead at AWS, serving because the International Tech Lead for Anthropic at AWS. He works on the intersection of cloud structure and Synthetic Intelligence, serving to organizations undertake and scale Anthropic’s applied sciences on AWS.
Antonio Rodriguez
Antonio Rodriguez is a Principal Specialist Options Architect for Amazon Bedrock at AWS, specializing in enterprise generative AI structure and controlled business deployments.
