Introducing Claude Fable 5.1 on AWS
Amazon Web Services Dani Mitchell ● Covered by 3 sources
AWS added Claude Fable 5.1 to Bedrock and Claude Platform on AWS. It’s pitched for hard coding and research work, but it also comes with tighter data rules.
Based on reporting by Amazon Web Services, Dani Mitchell — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
AWS says Claude Fable 5.1 is now available on Amazon Bedrock and on Claude Platform on AWS. The pitch is simple enough: this is Anthropic’s newest model for messy, high-stakes work — coding, scientific research, and enterprise jobs that don’t fit neatly into a prompt-and-pat it flow.
Anthropic is calling Fable 5.1 a Covered Model, which means extra retention, safety review, and access controls wherever it’s offered. On Bedrock, that means prompts and outputs can be retained for up to 30 days and reviewed by Amazon personnel. To use it, customers need aws_review mode, set before the model is invoked.
AWS and Anthropic are also pushing Enterprise Frontier Safeguards, or EFS. For eligible customers, the setup allows Claude Fable 5 and 5.1 to run with zero data retention on Amazon Bedrock and Claude Platform on AWS, with internal use available through December 31, 2026. Later this year, AWS says customers will also be able to keep prompts and outputs in their own AWS account, under their own encryption keys, access policies, and audit logs, with automated review instead of human review.
The model itself is being sold as better at the ugly parts of AI work. Anthropic says Fable 5.1 improves on Fable 5 on hard reasoning tests, including competition math, graduate-level engineering and science questions, and long multi-step problems where reliability matters. It’s also supposed to be better at agentic coding, multi-hour autonomous work, end-to-end knowledge work, and scientific research campaigns.
Getting access is straightforward if you already live inside AWS. You can try it in the Bedrock console under Test > Playground, call it through the Anthropic Messages API on bedrock-runtime, or use Invoke and Converse through the AWS CLI and SDKs. AWS says Fable 5.1 is available today through US Geo CRIS and Global CRIS inference profiles, and in AWS GovCloud (US) on bedrock-runtime and bedrock-mantle. It’s also available through Claude Platform on AWS in North America.
My take — AI-written commentary, not fact-checked reporting
This is AWS doing what AWS does best: turning a model launch into a governance story with a console button attached. The real headline isn’t just that Fable 5.1 is capable; it’s that the default conversation around powerful models is now retention, review, encryption keys, and who gets to look at the prompts. That’s the market maturing, whether the hype crowd likes it or not.
Read more about this at: Amazon Web Services
