Anthropic introduced Claude Haiku 5.5 on October 7 for high-volume work such as summaries, classification and database queries. The company positions it as a lower-cost subagent alongside Opus 5.5 and Sonnet 5.5 on coding tasks.
Haiku 5.5 is the first Haiku model with an adjustable effort setting. A developer can choose how much reasoning it spends on a request, trading additional work against response time and cost. Anthropic says the model costs around 75% less to run than Haiku 4.5 on average; the saving depends on the workload and selected effort.
The published tests cover computer use, coding and professional tasks. On FrontierCode 1.1, which evaluates code changes, Anthropic reports 46.4% for Haiku 5.5 against 52.1% for Sonnet 5.5 at its listed setting. The larger model still has an advantage on that test.
The launch also halves Sonnet 5.5’s cache-read price and introduces a monthly API credit for Claude Max and Team subscribers.