Anthropic · Text generation
Claude Haiku 5.5
Claude Haiku 5.5 is designed for high-volume, latency-sensitive work. Use it to classify support messages, extract fields from supplied documents, route requests, summarize short passages or explain a screenshot. Text and image understanding support concise, well-scoped answers; selectable effort trades speed for more deliberate reasoning. Medium is the balanced default, while Low suits simple repetitive tasks. For ambiguous, long-horizon analysis, compare a larger reasoning model. Flowctory returns text and does not expose autonomous computer, browser or hosted-search tools for this implementation.
Generation mode
Text generation
Credit cost
Priced by text length
Observed generation time
Not enough public data
Release date
Oct 7, 2026
Current credit pricing
Text generation credits depend on token usage and the selected settings.
Text model capabilities
- Maximum output tokens
- 128,000
- Image input
- Supported
- Video input
- Not supported
- Web search
- Not supported
- Reasoning
- Supported
Availability and credit rows come from Flowctory's active production catalog. Observed time is shown only after at least five completed generations and is not a controlled benchmark. Model availability, limits, and costs can change.
Use model