Anthropic launches Claude Haiku 5.5, the first Haiku model with effort controls, for high-volume, cost-sensitive tasks like summaries and classification

The cost to process millions of tokens via Anthropic's Haiku model falls by up to 75%.

Anthropic has launched Claude Haiku 5.5, an updated version of its smallest and most cost-effective language model. This new iteration is designed for high-volume, cost-sensitive applications such as data summarization, classification, and database queries. A key advancement is the introduction of adjustable effort controls, allowing users to balance cost and intelligence based on their specific needs, a feature previously only available in larger models. Anthropic reports that Haiku 5.5 costs approximately 75% less to operate than its predecessor, Haiku 4.5. The company also announced price reductions for its Claude Sonnet 5.5 model's cache reads, making it about 20% cheaper for agentic tasks. New monthly API credits will be provided to Max and Team subscribers to encourage development on the Claude Platform. Haiku 5.5 is available on major cloud platforms including AWS, Google Cloud, and Azure.

The release of Claude Haiku 5.5 signifies Anthropic's push to make advanced AI capabilities more accessible for routine, high-throughput tasks. By integrating adjustable effort controls into their smallest model, Anthropic empowers users to fine-tune performance versus cost for specific use cases, a crucial factor for developers managing large-scale applications. This move not only democratizes access to sophisticated AI but also sets a new benchmark for cost-efficiency in the market, potentially driving broader adoption of AI agents for everyday business processes.

This strategic pricing and performance adjustment will likely impact businesses relying on high-volume AI tasks, enabling them to significantly reduce operational expenses. The improved cost-efficiency of Haiku 5.5, coupled with the reduced pricing for Sonnet 5.5 cache reads, suggests a market trend towards specialized, cost-optimized AI solutions. Developers and businesses should evaluate how these changes can enhance their existing workflows or enable new, previously cost-prohibitive applications.