Claude Haiku 5.5
Claude Haiku 5.5 is Anthropic's fastest current model, released October 7, 2026 with the API ID claude-haiku-5-5.
Provider
Anthropic
Model family
Anthropic Claude
Multimodal LLM
Cost tier
Haiku
Status
Current
Release Oct 7, 2026
Technical specs
- Inputs
- text, image
- Outputs
- text
- License
- Proprietary API
- Latency
- Fastest
- Context tokens
- 1,000,000
- Default Effort
- medium
- Knowledge Cutoff
- 2026-06
- Max output tokens
- 128,000
- Input Usd Per Million
- 0.1
- Output Usd Per Million
- 0.5
- Cache Read Usd Per Million
- 0.01
- Price Tier Threshold Tokens
- 100,000
- Max Output Tokens Batch Beta
- 300,000
- Cache Write 1h Usd Per Million
- 0.2
- Cache Write 5m Usd Per Million
- 0.125
- Long Prompt Input Usd Per Million
- 0.5
- Long Prompt Output Usd Per Million
- 2.5
- Long Prompt Cache Read Usd Per Million
- 0.05
- Long Prompt Cache Write 1h Usd Per Million
- 1
- Long Prompt Cache Write 5m Usd Per Million
- 0.625
- Capabilities
- 1 million token context window
- 128K max output, 300K on Batch API (beta)
- Adaptive thinking with effort control (default medium)
- Available on Claude API, Amazon Bedrock, Google Cloud, Microsoft Foundry, and Claude Platform on AWS
Why teams choose it
Complex reasoning
Useful for workflows that require structured thinking, multi-step logic, and deeper analysis than lightweight models provide.
Long-context analysis
Helps teams summarize, compare, and extract insights from long documents without losing important nuance.
Coding and tools
Works well for code assistance, tool calling, and agent workflows where instructions must stay consistent across steps.
Cost-efficient routing
Useful as part of a routing stack where cheap models handle drafts and confirmations and this tier handles genuinely hard passages.
Tradeoffs to know
- Prices rise five-fold for prompts over 100,000 tokens.
- The newer tokenizer counts about 30% more tokens than Claude Haiku 4.5 for the same text.
- temperature, top_p, and top_k should be omitted; some values return a 400 error.
Benchmarks
No independent or vendor benchmark scores in this catalog entry.
Anthropic Claude family lineup
Current models
Previous versions
Related models
Claude Haiku 5.5 FAQ
What is Claude Haiku 5.5?
5 is Anthropic's fastest current model, released October 7, 2026 with the API ID claude-haiku-5-5. Anthropic positions it for high-volume, latency-sensitive work such as classification, routing, extraction, and subagent tasks. It takes text and images and returns text, with a 1 million token context window, 128K max output (300K on the Message Batches API with the output-300k-2026-03-24 beta header), and adaptive thinking with a default effort of medium. 05 cache reads. 5.
When does Claude Haiku 5.5 fit best?
High-volume classification, extraction, and routing
What should teams watch out for with Claude Haiku 5.5?
Prices rise five-fold for prompts over 100,000 tokens.
Explore next
Models, tools, and comparisons that connect to this reference.