David Corvoysier

@baggiponte thank you for your feedback. As you can see in the commit below, the only reason there is a `setup.py` is because of an issue when parsing metadata using...

Yes (the previous commit). (Completely off-topic, but I have the exact same desk as yours: ikea ?).

> I noticed there are smooth quant implemented under [external](https://github.com/huggingface/quanto/tree/main/external/smoothquant). Currently, its implementation seems to be model-specific, we can only apply smooth on special Linear. This is a copy/paste from...

@mattiadg sure: feel free to select a model and create an example.

@mattiadg yes a whisper example would be awesome

Nice ! Looking forward to see your example. When submitting your pull-request, please do not forget to use [conventional commits](https://www.conventionalcommits.org/en/v1.0.0/) to make it easier to review.

> It seems the `QLayerNorm` doesn't have a quantization weight type. Is this a issue with `QLayerNorm`? Yes, line 168 should be modified to support `weight_qtype == None`, as the...

Do it in the same pull-request, but in a separate commit. They are not squashed when merging so it will appear as a stand-alone change.

cc @younesbelkada @SunMarc

I have made some progress on this ([branch](https://github.com/huggingface/quanto/tree/awq_kernels)): - I verified that I could replicate AWQ packing format, - I added a method to unpack AWQ packed tensors (required for...