Haiku 5.5 cuts short-prompt prices and adds effort controls

Anthropic released Claude Haiku 5.5 on 7 October for high-volume tasks such as summaries, classification and smaller agent assignments. The announcement priced prompts up to 100,000 tokens at $0.10 per million input tokens and $0.50 per million output tokens.

Above that prompt threshold, the respective rates rise to $0.50 and $2.50. Those are separate input and output charges, not a flat price for a task. The threshold makes context management relevant to a product’s operating costs.

Haiku 5.5 is the first Haiku model with adjustable effort, letting users trade additional reasoning against cost. Anthropic made it available through its developer platform and major cloud providers and positioned it alongside larger models that can assign narrower jobs to it. The company’s claimed capability gains are vendor evaluations. The concrete launch gives application developers a cheaper short-prompt option with controls that previously belonged to larger models. Developers can evaluate that option against the narrower jobs they actually run.