Anthropic released Claude Opus 5.5 on AWS, saying it performs more with fewer tokens than its predecessor, improving efficiency for agentic coding and long-running tasks. It is the company's first major update to the Opus model family since the launch of Claude Opus 5.
AWS reported lower per-token prices and cheaper cache reads for Claude Opus 5.5, measured on task cost efficiency. That compares with higher costs for Claude Opus 5, allowing teams to run more ambitious workflows at scale.
Claude Opus 5.5 is built on AWS Bedrock and targets industries requiring consistency and depth, such as software development and knowledge work. Availability begins today, initially for users on Amazon Bedrock and Claude Platform on AWS.
"Claude Opus 5.5 does more with fewer tokens than Claude Opus 5, and new pricing passes those gains straight to customers," said Anthropic. The model's efficiency gains allow teams to run more ambitious agentic work at scale.
The announcement follows the release of Claude Fable 5.1, which introduced safety classifiers for specific domains. Anthropic emphasized the importance of safety and efficiency in its latest model update.
Anthropic did not say how the model will perform on specific benchmarks, and the company raised questions about the trade-offs between cost and performance. The model is now available for testing through the Amazon Bedrock console.
Source: awsml
