Amazon Web Services (AWS) has expanded its range of artificial intelligence (AI) models available through Amazon Bedrock by adding Claude Haiku 4.5, joining the previously launched Claude Sonnet 4.5 and Claude Opus 4.1 from Anthropic. These models are designed to offer hybrid reasoning capabilities, allowing users to choose between fast responses and deeper, more extended reasoning for complex tasks.
The Claude 4 models aim to improve how businesses use AI in both high-volume operations and multi-step workflows. Claude Haiku 4.5 is positioned as Anthropic’s most efficient model, delivering fast and cost-effective performance suitable for scaled sub-agent deployments and applications where quick response times are essential, such as customer service agents and chatbots.
Claude Opus 4.1 brings enhanced precision and performance over its predecessor in coding and agentic tasks. It is capable of independently planning and executing complex development projects, adapting to user styles, and handling long-horizon tasks with accuracy.
Claude Sonnet 4.5 is described by Anthropic as their most intelligent model, excelling in advanced reasoning, coding, and coordination with sub-agents. Kate Jensen, head of Growth and Revenue at Anthropic, stated: "Claude Sonnet 4.5 transforms AI from a tool into a true collaborator for every person and every team. Our customers will see project timelines shrink—in many cases from weeks to hours. Claude models consistently set new standards in coding, advanced reasoning, and multi-step workflows while understanding full business contexts and delivering precise results. The real breakthrough is freeing your talent for strategic work while Claude handles the heavy lifting."
All Claude 4 models feature “extended thinking,” enabling them to switch between providing quick answers and step-by-step reasoning. This flexibility is intended to support a wide range of use cases from real-time customer support to complex research or software development.
Anthropic reports that Claude Sonnet 4.5 has achieved a coding performance score of 77.2% on the SWE-bench benchmark, increasing to 82% when evaluating multiple answers at once. Claude Haiku 4.5 scored 73.3% on the same benchmark, indicating its utility for parallelized execution and sub-agent applications.
The addition of these models to Amazon Bedrock gives customers access to more advanced AI tools while maintaining enterprise-grade security and responsible AI controls.