Implementing LanguageModelV1 Interface needed?

Hello people. I need a push in the right direction because I think I am doing something wrong here, or misunderstanding it.

Currently I am using the AI Chatbot template. It gives great functionality and I wish to use it. However I am stuck with implementing my own Model.

Actual Question:
I need to add a new custom model which I have somewhere in a remote URL. This model just streams back text. Do I need to implement a customProvider for this as described here:

Or am I missing something completely? I am specifically talking about the backend code see below.
I did a shitty frontend where as POC where the stream back works but I cant get it to integrate with the provided structure of the vercel ai sdk.

Any tips or pushes in the correct direction would be greatly appreciated

Hi, David! Welcome to the Vercel Community :wave:

I admittedly haven’t played around with the AI SDK as much to add many comments. However, I did ask v0 for some pointers → Vercel AI Custom Model Integration – v0 by Vercel

Let me know if that helps!