Skip to main content

System status

Coverage is stale.

Collection is paused. Latest public event: Aug 21, 2026 (10 days ago).

← Intel index

This coverage is stale.

Last updated May 21, 2026 (about 3 months ago).

product · May 21, 2026

LiteLLM Releases Experimental Microservices Helm Chart for Scalable Deployment

Share the canonical public link.

Share as image

LiteLLM introduced experimental microservices Helm chart in version 1.86.0-dev to split proxy into three independently scalable services. The gateway service handles LLM data plane on port 4000, backend manages management and UI API on port 4001, and static UI runs on port 3000 with a one-shot migrations Job. Chart supports HPA per component with defaults such as gateway maxReplicas 10 and requires Kubernetes cluster plus external Postgres. This deployment isolates inference traffic from control-plane slowdowns compared to prior single-image setup.

Below validation threshold — auto-passed without scoring

Supporting evidence