/kind feature

Describe the solution you'd like

I would like to use storage initializer to pull the huggingface models which fits the KServe experience. StorageUri will be in this format: hf://<repo>/<model>:<hash(optional)>

apiVersion: serving.kserve.io/v1beta1
kind: InferenceService
metadata:
  name: bloom
spec:
  predictor:
    model:
      modelFormat:
        name: huggingface
      #runtime: kserve-huggingface
      storageUri: hf://bigscience/bloom-560m:<hash>
      resources:
        limits:
          cpu: "2"
          memory: 4Gi
        requests:
          cpu: "1"
          memory: 2Gi

Anything else you would like to add:

Another option is to support any Git LFS.

Links to the design documents:
[Optional, start with the short-form RFC template to outline your ideas and get early feedback.]
[Required, use the longer-form design doc template to specify and discuss your design in more detail]