v1.101.0 v1.102.0
BerriAI/litellmSep 15, 2026 → Sep 22, 20268 days between releases
What's New
Adds streaming guardrails, Claude Code integration, and new audio/embedding models
New Features
- Execute post_call guardrail pipelines on streaming responses
- Enhanced Claude Code integration via 'lite pi' command
- Meta Muse Voice transcription support
- TwelveLabs embeddings support
- MCP schema discovery proxy mode and OAuth relay
Improvements
- Proxy redaction fix for sensitive data
- GovCloud pricing support for Bedrock models
Bug Fixes
- Apply post_call pipeline text rewrites on streams
- Proxy redaction fix for sensitive data
Upgrade Notes
Guardrail pipelines now execute on streaming responses, which may impact latency. Verify Claude Code integration settings if using the new 'lite pi' command. No configuration changes are required for the new audio or embedding models.
The Two Releases
v1.101.0
Sep 15, 2026· 1 month ago
This release introduces Docker image signature verification using cosign for enhanced security and includes numerous bug fixes across proxy, bedrock, and guardrails components, alongside new features for OpenAI workload identity federation and gigachat passthrough routing.
Full release notesv1.102.0
Sep 22, 2026· 29 days ago
This release focuses on enhancing the LiteLLM proxy with robust guardrail capabilities, improved routing strategies, and expanded model support including new real-time audio features. It introduces significant updates to the UI, MCP (Model Context Protocol), and enterprise deployment tools like Helm and Terraform. Additionally, the release emphasizes security by signing all Docker images with Cosign.
Full release notes