v1.84.9 - Anthropic Cache-Control Cap
Deploy this version​
- Docker
- Pip
docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
docker.litellm.ai/berriai/litellm:1.84.9
pip install litellm==1.84.9
v1.84.9 is a patch release on top of v1.84.8. It caps Anthropic cache_control injection at the 4-block API limit so prompt-caching requests no longer fail when more blocks are eligible.
What's Changed​
- fix(integrations): cap Anthropic cache_control injection at 4 blocks - PR #30480
Full Changelog​
https://github.com/BerriAI/litellm/compare/v1.84.8...v1.84.9