Michał Krassowski

This also affects other `List` traits with `allow_none=True`, namely `--AiExtension.allowed_models` and `--AiExtension.blocked_models`. I tested that `Dict` traits work fine on `--AiExtension.model_parameters`.

I think that maybe we can fix it in jupyter-ai by specifying `allow_none = False` and adding a `default=[]`. I will try it another day. I agree no need to...

> `traitlets==5.13.0` was released on PyPI on October 30, 2023 Yes, one week after https://github.com/jupyterlab/jupyter-ai/pull/415 was merged. And on November 8th the test was disabled in https://github.com/jupyterlab/jupyter-ai/pull/446, which I suspect...

It is possible to add traitlet configuration to magics, in IPython core this is done for `%store` magic [here](https://github.com/ipython/ipython/blob/f22a925d565fa018d41d6720de8bf4df85e797c7/IPython/extensions/storemagic.py#L66-L70) and then user just sets `c.StoreMagics.autorestore = True` in their config...

1. Does chat work with the same model? 2. Does completion work with a different, non-local model? 3. Does setting completion streaming to "always" help? My first guess would be...

Also, are you using the latest JupyterLab version?

From your pictures it looks like you are expecting the completion to appear in a new line (based on the position of your cursor). Currently it works by completing the...

My question is if chat model with the same exact model works, not with a different model. Conversly, if you try using Gemma for inline completion does it work for...

If other models work but starcoder does not work, it is likely a problem with the model, or in this case also possibly your GPU having less memory than required...

If you use streaming mode, this should have been fixed in https://github.com/jupyterlab/jupyter-ai/pull/879 (which will be included in 2.19 release)