Model · Anthropic · Text
Claude Haiku 5.5
Anthropic's fastest, cheapest tier — built for high-volume, latency-sensitive work such as classification, extraction, routing, and subagent tasks. Succeeds Claude Haiku 4.5.
- Modality
- Text
- License
- Proprietary (Proprietary)
- Context window
- 1,000,000 tokens 1M tokens is both the default and the maximum. Up to 128k output, or 300k on the Batch API with the output-300k beta header. Pricing is tiered by prompt length: prompts over 100k tokens bill at 5x the base rate.
- Released
- October 7, 2026
- Last verified
- October 28, 2026
- Runs locally
- No
- Also handles
- vision
Strengths
- Fastest model in the Claude lineup
- Very cheap on short prompts: $0.10/$0.50 per MTok for prompts up to 100k tokens
- 1M-token context and 128k output — a big jump from Haiku 4.5's 200k context
- Adaptive thinking with the effort parameter (default `medium`)
Weaknesses
- Prompts over 100k tokens cost $0.50/$2.50 per MTok, five times the short-prompt rate
- Newer tokenizer counts the same text as roughly 30% more tokens than Haiku 4.5
- Thinking blocks only work in the account that produced them, or a linked account
- `temperature`, `top_p`, and `top_k` must be omitted
- Closed weights — cannot be self-hosted
Try it
| Where | Type | Notes |
|---|---|---|
| Anthropic API | hosted-api | API key required |
| AWS Bedrock | hosted-api | Enterprise / regional |
| Google Vertex AI | hosted-api | GCP integration |
Used in solutions
Version history
- Claude Haiku 5.5 Oct 2026 Current
- Claude Sonnet 5.5 Sep 2026
- Claude Opus 5.5 Sep 2026
- Claude Fable 5.1 Sep 2026
- Claude Mythos 5.1 Sep 2026
Official sources
- Model docs docs
- What's new in Haiku 5.5 docs
- Models overview docs
Change log
- — Initial entry. Released 2026-10-07, superseding Claude Haiku 4.5; $0.10/$0.50 per MTok for prompts up to 100k tokens ($0.50/$2.50 above), 1M-token context, 128k output.
Esc