v1.88.4 - Proxy Exception & Guardrails Fixes
Deploy this version​
- Docker
- Pip
docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
docker.litellm.ai/berriai/litellm:1.88.4
pip install litellm==1.88.4
v1.88.4 is a patch release on top of v1.88.3. It restores readable ProxyException messages, returns 400 instead of 500 when AIM guardrails block a request, caps Anthropic cache-control injection, and corrects Datadog batch splitting and a chat-completions flag leak.
What's Changed​
- fix(proxy): populate Exception.args so str(ProxyException) returns message - PR #29015
- fix(datadog): split oversized batches on 413 instead of re-queueing forever - PR #29444
- fix: stop use_chat_completions_api flag from leaking into provider request body - PR #29447
- fix(integrations): cap Anthropic cache_control injection at 4 blocks - PR #30480
- fix(guardrails): return 400 not 500 when AIM blocks a request - PR #30573
Full Changelog​
https://github.com/BerriAI/litellm/compare/v1.88.3...v1.88.4