Claude Haiku 5.5
- API model IDs
- claude-haiku-5-5
- Context window
- 1,000,000 tokens
- Max output
- 128,000 tokens
- Released
- 2026-10-07
- Lifecycle
- active, retires not sooner than 2027-10-07
- License
- proprietary
Lifecycle verified 2026-11-05 (0 days ago). Source
Data handling
- Trains on API data by default
- No
- Retention
- 30 days
- Zero data retention
- on-request
- Residency
- global, us
- HIPAA BAA
- Available
- SOC 2 / FedRAMP
- SOC 2 yes, FedRAMP none
Anthropic's published certifications list for commercial services names ISO 27001:2022, ISO/IEC 42001:2023, and SOC 2 Type I and Type II. It does not name a FedRAMP authorization for the first-party Claude API.
Verified 2026-09-08 (58 days ago). Source 1, Source 2, Source 3, Source 4, Source 5 Subprocessors
Pricing snapshot
$0.1 input, $0.5 output, per million tokens. Cached input $0.01 per million tokens. Batch requests are 50% off.
This is a snapshot verified on 2026-11-05, 0 days ago, not a live price. Confirm on the provider’s pricing page before you budget against it.
Where you can call it
- direct · claude-haiku-5-5
- bedrock · anthropic.claude-haiku-5-5
- vertex · claude-haiku-5-5
- azure · claude-haiku-5-5 Microsoft Foundry deployment name.
Priced by prompt length. The price shown is for prompts up to 100,000 tokens. A request whose prompt is over 100,000 tokens, counting cache reads and writes, bills at $0.50 input, $0.05 cache read, and $2.50 output per million tokens instead of $0.10, $0.01, and $0.50. It is the only current Claude model without flat pricing across its 1M token context window. It uses the tokenizer introduced with Claude Opus 4.7, which Anthropic says produces approximately 30 percent more tokens than Claude Haiku 4.5 for the same text. On the Message Batches API it supports up to 300,000 output tokens with the output-300k-2026-03-24 beta header.