Last released Sep 23, 2026
Self-hosted LLM inference with a control plane, load balancer and OpenAI-compatible API