TLDR: Amazon has announced the immediate availability of Anthropic’s Claude Haiku 4.5 within its Amazon Bedrock generative AI service. This integration expands Bedrock’s offering of Claude 4 models, providing customers with Anthropic’s most efficient and cost-effective model, optimized for speed and performance in critical, real-time AI applications.
Amazon has officially announced the integration of Anthropic’s Claude Haiku 4.5 into its Amazon Bedrock generative AI service, effective October 15, 2025. This addition significantly bolsters Amazon Bedrock’s suite of advanced AI models, which already includes Claude Sonnet 4.5 and Claude Opus 4.1.
Claude Haiku 4.5 is distinguished as Anthropic’s most efficient model, offering near-frontier performance with remarkable speed and cost-effectiveness. This makes it an ideal choice for scaled sub-agent deployments where optimal intelligence and rapid response times are paramount. Amazon highlighted that the model’s exceptional speed will make AI agents, such as customer service representatives and chatbots, faster and more responsive, which is crucial for use cases demanding low-latency interactions.
Anthropic’s latest offering is part of a broader family of ‘hybrid reasoning models’ that can seamlessly switch between near-instant responses and more extended, thoughtful processing. These models are setting new benchmarks in areas like coding, advanced reasoning, and multi-step workflows. While Claude Sonnet 4.5 and Claude Opus 4.1 are designed for sustained performance on complex, long-running tasks and can power AI agents capable of completing hours of work in minutes, Claude Haiku 4.5 excels in scenarios requiring quick, efficient processing.
Amazon Bedrock, a fully managed generative AI service from Amazon Web Services (AWS), provides organizations with a unified API to access, experiment with, and deploy leading foundation models from various providers, including Anthropic, AI21 Labs, Cohere, Meta, Mistral AI, Stability AI, and Amazon itself. This service is designed to empower developers and enterprises to build, customize, and scale generative AI applications securely and efficiently, eliminating the need for extensive server or machine learning infrastructure management.
Beyond Amazon Bedrock, Claude Haiku 4.5 is also available on Claude.ai, through the Anthropic API, and on Google Cloud’s Vertex AI. Anthropic has priced Claude Haiku 4.5 at $1 per million input tokens and $5 per million output tokens, a slight increase from Haiku 3.5’s $0.80/$4 per million input/output tokens, reflecting its enhanced capabilities.
Also Read:
- Key AI Leadership Shifts at AWS Amid Intensified Talent Competition
- AWS Enhances Cloud Financial Management Training with Dedicated FinOps for Generative AI Course
Despite its smaller size, Claude Haiku 4.5 has demonstrated performance on par with Claude Sonnet 4 in coding tasks, achieving this at one-third the cost and more than twice the speed. It can even surpass Sonnet 4 in computer usage tasks. Developers can leverage Haiku 4.5 in conjunction with the more powerful Claude Sonnet 4.5, where Sonnet can devise multi-step plans, and a ‘team of multiple Haiku 4.5s’ can then execute subtasks in parallel. Additionally, Claude for Enterprise premium seats with Claude Code, which includes Claude Sonnet 4.5 and Haiku 4.5, are now available in AWS Marketplace, offering flexible integration options for enterprise teams to embed AI-powered development capabilities into their workflows.
