Adding model does not check remote model access/authorization

[strus38]

Hi,

Environment: kubeai:0.4.1

Difficult to write the right title for this one, but basically I added a model thanks to the new documentation provided last night - thank you!

apiVersion: kubeai.org/v1
kind: Model
metadata:
  name: llama-3-8b
spec:
  features: ["TextGeneration"]
  owner: meta-llama
  url: hf://meta-llama/Meta-Llama-3-8B
  engine: VLLM
  minReplicas: 0
  maxReplicas: 3
  resourceProfile: CPU:4
  env:
      VLLM_CPU_KVCACHE_SPACE: "4"
  args:
  - --max-model-len=32768
  - --max-num-batched-token=32768

Then calling this model through the openwebui, it is stucked.
Looking at the kube logs on the pod model, I see:

...
huggingface_hub.utils._errors.GatedRepoError: 403 Client Error. (Request ID: Root=1-66d1c249-75919cc37bdd839c3d8506ba;c3e0ae52-ae6f-4463-9192-5c53d581fbfc)

Cannot access gated repo for url https://huggingface.co/meta-llama/Meta-Llama-3-8B/resolve/main/config.json.
Access to model meta-llama/Meta-Llama-3-8B is restricted and you are not in the authorized list. Visit https://huggingface.co/meta-llama/Meta-Llama-3-8B to ask for access.

Fine, some models are protected, however, how could kubeai be improved in order to know that beforehands and not discovering that through the logs?
Is it more on openwebui side, or can something be done at kubeai level to raise this when model are added?
I would also accept the 'live with it and look at licences before using the models' answer.
Thanks!

[nstogner]

I agree we need to tackle this poor experience soon. We have heard this same feedback on the Discord channel as well.