v1.99.3 - Claude Code Auto Mode and TypeSafe Jev Passthrough
Deploy this version​
- Docker
- Pip
docker run \
-e LITELLM_MASTER_KEY=sk-<paste-a-long-random-key> \
-e DATABASE_URL=postgresql://<user>:<password>@<host>:5432/<dbname> \
-e STORE_MODEL_IN_DB=True \
-p 4000:4000 \
docker.litellm.ai/berriai/litellm:1.99.3
pip install litellm==1.99.3
This release is published as ghcr.io/berriai/litellm:v1.99.3. See the GitHub release and the full releases page
v1.99.3 is a patch release on top of v1.99.2. It brings Claude Code auto mode through the gateway and adds the TypeSafe Jev passthrough. There are no new database migrations or breaking changes. The v1.99.3 tag points at 16923e0
Claude Code auto mode works through the gateway​
Claude Code auto mode sends a safeguards field on /v1/messages and expects safeguard_results in the reply. Earlier releases dropped the field, so /status showed Auto mode server: Disabled and every tool call fell back to the client-side classifier
The proxy now forwards safeguards and anthropic-beta unchanged on native Anthropic /v1/messages, and forwards safeguards plus the dangerous-tool-use-2026-09-03 beta to Bedrock Invoke and Vertex AI
TypeSafe Jev passthrough​
Set TYPESAFE_API_KEY and the proxy forwards /typesafe/* requests to TypeSafe, logs them, and tracks their spend. Pricing entries are added for typesafe/jev-latest, typesafe/jev-preview and typesafe/jev-1.13.0
What's Changed​
- fix(anthropic): forward safeguards and anthropic-beta unchanged on native /v1/messages -
9f6a6f8 - fix(anthropic): forward Claude Code safeguards and dangerous-tool-use beta to Bedrock Invoke and Vertex on /v1/messages -
a631bf7 - feat(typesafe): add TypeSafe Jev passthrough with logging and cost tracking -
7d1db43
Full Changelog​
https://github.com/BerriAI/litellm/compare/v1.99.1...v1.99.3
