What changed: Claude launched Claude Haiku 5.5 (claude-haiku-5-5), tuned for high-volume, latency-sensitive work, with a 1M token context window, 128k max output tokens, and adaptive thinking controlled via the effort parameter. It is available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry. Code written for Claude Haiku 4.5 can break on Haiku 5.5: manual extended thinking (budget_tokens) now returns a 400 error, adaptive thinking is on by default so responses can start with thinking blocks, and the same text counts as more tokens.
Who is affected
Businesses using Claude Haiku models via the Claude API, Amazon Bedrock, Claude Platform on AWS, Google Cloud, or Microsoft Foundry, especially those migrating from Haiku 4.5 to Haiku 5.5.
What it means for your business
For businesses running Claude Haiku in production, this means existing Haiku 4.5 integrations need review before switching to Haiku 5.5, since manual thinking parameters and token-count behavior change.
What to do
Review the migration guide before moving Haiku 4.5 integrations to Haiku 5.5, remove calls that pass budget_tokens, and update code to handle adaptive thinking blocks and higher token counts.
When it takes effect
Available as of October 7, 2026, per the release notes.
Source: https://platform.claude.com/docs/en/release-notes/overview#october-7-2026
