NVIDIA Switchyard AI Routing Policy Policy

Third Party

Observe routing decisions without applying them

Configure the AI Routing Provider policy in observe_only mode. The policy asks Switchyard which model it would choose and records the decision, but it doesn’t change where requests go. Use this to check Switchyard’s decisions on real traffic before you let them route requests.

Prerequisites

  • A running NVIDIA Switchyard decision service that the data plane can reach.

  • The AI Routing Provider custom policy registered, and the data plane running with KONG_UNTRUSTED_LUA=on.

  • An existing AI Model for each target, with no policies attached, and an AI Model that clients call.

Set up the policy

Help us make these docs great!

Kong Developer docs are open source. If you find these useful and want to make them better, contribute today!