ode

This is a big improvement when using ollama. Hopefully it's merged soon :+1:

Was working earlier, but did `git pull`, did `npm install` for client and pulled the new docker image - now the same problem as above.

Noticed the same and didn't find a way to change it. I wonder if it comes from the model (Whisper I think was trained on 30 second clips). I guess...

Fighting with the same thing here. Tried giving permissions in every possible way and nothing works... Perhaps some sleep and tomorrow will be brighter :crossed_fingers:

Ran into this also. In src/render/scheduler.rs, change: `_marker: std::marker::PhantomData::default(),` to: `_marker: PhantomData::default(),` and also: `.map(futures::StreamExt::fuse)` to: `.map(StreamExt::fuse)`