Allow ExcalidrawAI to use local running server Ollama

[chiangs]

I prefer to use locally running models. And in my workflow, I use the CoPilot plugin that allows me to tap into a locally running Ollama server and select different models.

Would be great to mirror this functionality in Excalidraw.

Thanks!

[sammcj]

Came looking for this feature as well, would be great!

[chrisk-7777]

Is it documented what service is used currently for the AI side of things? I can't find it anywhere, but its possible I am overlooking it.

I want to validate what context is being sent when using AI tooling