Claude Haiku 5.5
Claude Haiku 5.5 is Anthropic's small, fast model for high-volume and cost-sensitive work: summarisation, subagents and browser use. It succeeds Claude Haiku 4.5 with stronger coding and computer use, and at $0.10 per million input tokens it is the cheapest way onto the 1M-token Claude generation.
Technical Specifications
Live data fetched from the OpenRouter model registry — always up to date.
What is Claude Haiku 5.5?
Claude Haiku 5.5 is the high-volume tier of Anthropic's 5.5 generation. It succeeds Claude Haiku 4.5 and targets the work that happens thousands of times rather than once: summarising, classifying, extracting, driving subagents and controlling a browser. Despite the small footprint it keeps the full 1M-token context window, so a long transcript or log still fits in a single call.
Features & Capabilities
Fast and Cheap
$0.10 per million input tokens - built for volume, not for occasional use.
1M Context
The same million-token window as Opus and Sonnet at a fraction of the price.
Subagent Ready
Cheap enough to run many parallel workers behind one expensive planner.
Better Coding Than 4.5
Coding and computer use both improved over Claude Haiku 4.5.
Text, Image and File Input
Reads screenshots and documents as well as text.
High-Volume Value
Summarising a thousand items costs cents rather than dollars.
Who Is Claude Haiku 5.5 For?
Summarisation at Scale
Condensing transcripts, tickets, reviews and logs by the thousand.
Subagents
Many small workers feeding one planner without blowing the budget.
Browser and Computer Use
Page reading, form filling and UI checks that run continuously.
Classification and Extraction
Turning messy text into structured fields quickly.
Claude Haiku 5.5 Pricing
- $0.10 per million input tokens, $0.50 per million output tokens
- Cheapest tier of the Claude 5.5 generation
- Run it from your Hoodgen free credits, no credit card required
- Higher allowances on Hoodgen AI paid plans
What Makes Claude Haiku 5.5 Different?
Haiku 5.5 is interesting less for what it can do alone and more for what it makes possible in numbers. At $0.10 per million input tokens you can run a fan-out of twenty summarisers, extractors or page readers and still spend less than a single flagship call. The upgrade over Haiku 4.5 lands exactly where that shape of work needs it: coding and computer use.
Keeping the 1M-token window in the cheap tier changes the arithmetic further. A long support thread, a full log dump or a large batch of records fits in one request, so you are not paying for chunking and reassembly logic in output tokens.
The honest limitation is reasoning depth. Haiku 5.5 will follow a clear instruction reliably and will not untangle a genuinely ambiguous problem the way Opus 5.5 does. The pattern that works best is a strong planner with Haiku workers underneath - and Hoodgen lets you hold both on one account.
Claude Haiku 5.5 Performance & Pricing
Haiku 5.5 is the entry point to the Claude 5.5 generation. It trades reasoning depth for a price that supports running it constantly.
What Can You Do With Claude Haiku 5.5?
Use Haiku 5.5 wherever the instruction is clear and the volume is high. It is the tier that turns an expensive idea into something you can afford to run on every item rather than a sample.
Best Use Cases
- Bulk summarisation of tickets or transcripts
- Extraction into JSON for a pipeline
- Language detection and routing
- Subagents behind a larger planner
- Continuous browser and UI checks
Claude Haiku 5.5 Privacy & Data Policy
Claude Haiku 5.5 is developed by Anthropic and served through OpenRouter. Prompts and completions are processed by Anthropic under Anthropic terms.
As with any third-party model, avoid sending passwords or personal data. Hoodgen AI keeps larger and free alternatives on the same account so a sensitive task can be moved without rebuilding the workflow.
Claude Haiku 5.5 vs Other Models
Frequently Asked Questions
Volume work, flagship context
Chat with Claude Haiku 5.5 on Hoodgen - sign in and start in seconds.
Chat with Claude Haiku 5.5