Claude Opus 5.5 Now Available on AWS
AWS has announced the availability of Claude Opus 5.5 on Amazon Bedrock, offering improved efficiency, clarity, and safety features.
Today, AWS announced the availability of Claude Opus 5.5 on Amazon Bedrock and Claude Platform on AWS. This marks the introduction of the first model in the Claude 5.5 family, designed by Anthropic to be more capable for agentic coding, knowledge work, and long-running tasks.
According to Anthropic, Claude Opus 5.5 offers significant improvements over its predecessor, Claude Opus 5. It achieves more with fewer tokens, leading to lower per-token prices and cheaper cache reads. This results in an average lower cost per task compared to Claude Opus 5, enabling teams to undertake more ambitious agentic work at scale.
Claude Opus 5.5 is trained to communicate more clearly, providing updates on its actions, findings, and requirements during long-running tasks. Adaptive thinking is always active, allowing the model to determine the necessary level of reasoning for each task based on effort rather than manual thinking budgets.
Safety classifiers similar to those in Claude Fable 5.1 are included in Claude Opus 5.5, covering areas such as biology, cybersecurity, and AI development. Requests are more frequently refused compared to previous Opus versions, enhancing safety.
Use cases for Claude Opus 5.5 include industries where consistency and depth are critical. In software development, it offers improvements over Opus 5 for longer sessions with clear communication and explainability. For knowledge work, it requires fewer corrections while handling and creating long documents and reports.
To get started with Claude Opus 5.5, users can access the Amazon Bedrock console, choose Test, then Playground, and select Claude Opus 5.5 as the model. From there, prompts can be run directly against the model.
Source: aws-ml