Hey everyone, I've written a short doc to implement the Hugging Face model handler. This would enable loading models from Hugging Face directly and use it in RunInference. I've discussed a couple of approaches from users' point of view in the doc, feel free to comment and suggest what you think.
Thanks! Doc Link - https://docs.google.com/document/d/107T71QHMxvu67PyG-mkHwRhm7W2-07jihnqUMz-yuEM/edit?usp=sharing
