Inference Gateway
3 min read
Overview
This section covers the hands-on deployment procedure for a kgateway·Bifrost-based 2-Tier inference gateway. It includes Helm installation, HTTPRoute configuration, OTel integration, and troubleshooting.
Gateway concepts and routing strategy have moved to the Model Serving section
The gateway layer definitions, routing strategy, Cascade tuning, and OpenClaw implementation examples have moved to the Model Serving & Inference Infrastructure section so they can be read alongside the inference-infrastructure narrative. For the overall structure see Inference Infrastructure Overview, for the layer definitions see Tiered Gateway Architecture, and for the routing strategy see Routing Strategy.