Nvidia/Switchyard intelligent model routing setup

[enpalco]

Hello,

Has anybody tried to configure intelligent model routing using nvidia/switchyard in Openrouter on Goose agent. Details here https://openrouter.ai/docs/guides/routing/routers/switchyard-router.

If yes, can you share what was your setup and did it work fine? Im interested in a cloud setup of nvidia/switchyard router since i cannot afford a local host setup (not enough compute/memory and storage). Meaning switchyard router itself is local hosted, but the LLM models are routed to the cloud API providers.

Thanks a lot in advance.