v1.78.5.rc.1
📦 litellmView on GitHub →
✨ 4 features🐛 3 fixes
Summary
This release introduces team-level rate limiting features and enhances guardrail capabilities with content masking and streaming support. It also includes several bug fixes addressing memory leaks and configuration handling.
✨ New Features
- Team level model-specific tpm/rpm limits added.
- Working key-level validation of tpm/rpm limit when assigned to a team.
- Support for service_tier in chat completion.
- Added content masking and streaming support to PANW Prisma AIRS guardrail.
🐛 Bug Fixes
- Fixed a memory leak by ensuring pass through routes are only added when the path does not exist.
- Fixed an issue in proxy_server.py to re-encrypt environment variables on config save and use the original value on decrypt error.
- Added imagePullSecrets to migrations-job.