So currently it looks like we are unable to send the custom llm endpoint any information via query param (due to the appending of "/chat/completions") or through body (due to it not being exposed). I was wondering if there is a way to do something like this (i was thinking of using the agent name but it felt sort of hacky)
Use Case:
- Building an agent per tenant using vapi dashboard
- Each agent needs to let the custom llm server know the tenant that's calling