v1.90.4 - Responses API Guardrail Coverage
Deploy this version​
- Docker
- Pip
docker run \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
docker.litellm.ai/berriai/litellm:1.90.4
pip install litellm==1.90.4
v1.90.4 is a patch release on top of v1.90.3. It restores text guardrail coverage on the Responses API (/v1/responses). The shared guardrail content helpers now walk the Responses input taxonomy (the text, input_text, and output_text part types, and message, function_call, and function_call_output items), so guardrails that build on those helpers (AIM, Lakera v2, Cato, Lasso, Repello, IBM, Azure Content Safety, and enterprise secret detection) inspect and redact request text on /v1/responses the same way they already do for chat completions.
What's Changed​
- fix(guardrails): walk Responses-API text taxonomy in shared content helpers - PR #32542
Full Changelog​
https://github.com/BerriAI/litellm/compare/v1.90.3...v1.90.4