Function calling loading indictor

Hi there,

I’m wondering if someone has worked on implementing some kind of a loading indicator mechanism when the GPT response takes longer due to function calling. I’m mostly interested in how y’all do this, especially when streaming.

Basically I want a way to notify the frontend about the delay in response when GPT has chosen to use a function call. I’m using Vercel’s ai sdk with streaming

Any help / direction is highly appreciated!

1 Like