Claude Fable 5 on Google Cloud

Claude Fable 5 on Google Cloud is optimized for autonomous knowledge work and coding, handling long-running, complex, and asynchronous tasks.

Advanced AI Safety Notice: Claude Fable 5 on Google Cloud is within the scope of the Advanced AI Safety Addendum. All prompts and responses are stored for up to 30 days for the sole purpose of monitoring for potential abuse of relevant terms of service. See Responsible AI and Abuse Monitoring to learn more.

Retirement Date: Not sooner than June 8, 2027.

• Long-running agents: Run Claude Fable 5 on Google Cloud in an agent harness like Claude Code or Claude Managed Agents and it can work for days at a time: planning across stages, delegating to sub-agents, and checking its own work.
• Coding: Claude Fable 5 on Google Cloud is our most capable model for ambitious coding projects, including large migrations, complex implementations, and multi-day autonomous sessions. It can write its own tests to verify its work, implements designs with high fidelity, and uses vision to check outputs against goals.
• Enterprise workflows: Claude Fable 5 on Google Cloud handles complex, multi-stage knowledge work with minimal oversight, from deep research and analysis to finished deliverables. Teams can hand off large projects and review completed work rather than supervising every step.
• Financial analysis: Claude Fable 5 on Google Cloud brings deeper reasoning and best-in-class vision to financial workflows, reading dense filings, charts, and tables nested in PDFs at high fidelity. It supports investment research, earnings analysis, credit and risk review, and compliance workflows.
• Vision: Claude Fable 5 on Google Cloud understands diagrams, charts, and tables nested in files and PDFs, improving document-heavy work in finance, legal, analytics, and architecture. It also uses vision to verify its own work, checking outputs against the original design or goal.
• Computer use: Claude Fable 5 on Google Cloud is our most capable computer use model, pairing improved vision with deep reasoning for multi-step tasks that span multiple applications and require planning and judgment, including verifying and testing the same applications it creates.

Try in Agent Studio View model card in Model Garden

Model ID	claude-fable-5
Launch stage	GA
Supported inputs & outputs	Inputs: Text, Image, PDF Outputs: Text
Token limits	Maximum input tokens: 1,000,000 Maximum output tokens: 128,000
Capabilities	Supported Computer use Web search Batch predictions Prompt caching Function calling Count tokens Not supported
Usage types	Supported Shared Model Lineage Quota Provisioned Throughput Not supported
Technical specifications
	Images	Limitation and specifications: See Vision in Anthropic's documentation
	Documents	Limitation and specifications: See PDF support in Anthropic's documentation
Versions	claude-fable-5 Launch stage: Generally available Release date: June 9, 2026
Supported regions
	Model availability (Includes fixed quota & Provisioned Throughput)	United States Multi-region Europe Multi-region Global global endpoint
	ML processing	United States Multi-region Europe Multi-region Asia Pacific asia-southeast1
Quota limits	Multi-region: QPM: 1,000 Input TPM: 10,000,000 uncached and cache write Output TPM: 1,000,000 Context length: 1,000,000 Multi-region: QPM: 1,000 Input TPM: 10,000,000 uncached and cache write Output TPM: 1,000,000 Context length: 1,000,000 global endpoint: QPM: 2,000 Input TPM: 20,000,000 uncached and cache write Output TPM: 2,000,000 Context length: 1,000,000
Pricing	See Pricing.

Enable Claude Fable 5 on Google Cloud in your project

Follow these steps to enable Claude Fable 5 on Google Cloud in your project:

• Consent to the Advanced AI Safety Addendum in Model Garden (required once per project). See Abuse Monitoring to learn more.
• For online terms customers, click 'enable' on the Model Garden model card to:
  • Answer relevant company and use case questions in Model Garden.
  • Accept Anthropic and Marketplace terms of service in Cloud Marketplace.
• For Private Offer customers:
  • Accept the Private Offer in Cloud Marketplace (required once per Billing Account).
  • Enable the model location using the enableModel API
• As required by Section F of Anthropic's Service Specific Terms, enable prompt-response sharing with Anthropic for each model location. See Request-Response Logging and Sharing to learn more.