[Feature] Support AWQ quantization for HF AutoModel (embedding models)

[abhinavkulkarni]

Hi,

It should be possible to add AWQ quantization support for HuggingFace AutoModel which are used for generating embeddings.

Currently, AWQ support works for models of the type AutoModelForCausalLM.

Thanks!

[casper-hansen]

Hi @abhinavkulkarni, I would love to support more models - especially embedding models. If you are fortunate enough to have time to add support for these models, I would highly appreciate it