Claude Haiku 5.5
Anthropic's small, fast model at $0.10 per million input tokens — a 90% cut for short prompts.
Quick verdict
- Best for
- High-volume classification, extraction, and summarisation
- Pricing
- $0.10 / $0.50 per million tokens up to 100K-token prompts; $0.50 / $2.50 above
- Not ideal if
- Long prompts cost five times more — and the higher rate applies to the whole request
What is Claude Haiku 5.5?
Claude Haiku 5.5, released on October 7, 2026, completes the Claude 5.5 family as the model for high-volume work: summarisation, classification, subagents, and browser and computer-use tasks. Prompts up to 100K tokens cost $0.10 per million input and $0.50 per million output, against $1 and $5 for Haiku 4.5; prompts above 100K tokens bill at $0.50 and $2.50 for the whole request. It adds a configurable effort setting and beta computer-use and browser-use support in the SDKs.
Key features
- Two-tier pricing by prompt length (100K-token threshold)
- Configurable effort setting
- Beta computer-use and browser-use in the Python and TypeScript SDKs
- Anthropic-reported 72.4% on the OSWorld 2.1 offline subset (Haiku 4.5: 15.7%)
- Available on the Claude Platform and AWS Bedrock
Pros
- Roughly a 90% price cut for the short prompts that make up most traffic
- Effort control lets one model cover cheap and careful tasks
- Computer use at a budget price point
Cons
- Long prompts cost five times more — and the higher rate applies to the whole request
- Temperature, top_p, and top_k parameters now return errors; code from Haiku 4.5 needs updating
- Benchmarks are Anthropic's own
Best for
Read more
Related comparisons
Alternatives to Claude Haiku 5.5
DeepSeek V4.1 Flash
MIT-licensed open weights, native vision, a 1M context, and $0.15 per million input tokens off-peak.
GPT-6.1 Sol
OpenAI's efficiency model: near-Astra results at one-fifth of the token price.
Claude
Anthropic's AI — preferred by writers, developers, and researchers.