Automatic Prompt Caching

Claude's Messages API now supports automatic caching without requiring manual breakpoint management. Developers can add a single cache_control field to their request body, and the system automatically caches the last cacheable block while moving the cache point forward as conversations grow. This feature works alongside existing block-level cache control for fine-grained optimization and is available on the Claude API and Azure AI Foundry (preview).

Model Retirements and Upgrades

Anthropic has retired two models effective immediately:

• Claude Sonnet 3.7 (claude-3-7-sonnet-20250219) — upgrade to Claude Sonnet 4.6
• Claude Haiku 3.5 (claude-3-5-haiku-20241022) — upgrade to Claude Haiku 4.5

All requests to these models will now return errors. Additionally, Claude Haiku 3 (claude-3-haiku-20240307) has been marked for deprecation with retirement scheduled for April 19, 2026.

Researchers requiring ongoing access to retired models can request access through the External Researcher Access Program.

Action Items for Developers

Update your API requests to use the current model versions:

• Replace claude-3-7-sonnet-20250219 with claude-sonnet-4.6
• Replace claude-3-5-haiku-20241022 with claude-haiku-4.5
• Plan migration of claude-3-haiku-20240307 before April 19, 2026