bug: Access to restricted models like google/embeddinggemma-300m requires authorization(HF TOKEN)

[samzong]

Describe the bug

the CI run failed.

huggingface_hub.errors.GatedRepoError: 401 Client Error. (Request ID: Root=1-693b313c-3a360d925a720ea722d0d659;d80b0023-f1de-421c-828f-7f393a1763c2)

Cannot access gated repo for url https://huggingface.co/google/embeddinggemma-300m/resolve/57c266a740f537b4dc058e1b0cda161fd15afa75/.gitattributes.
Access to model google/embeddinggemma-300m is restricted. You must have access to it and be authenticated to access it. Please log in.
make[3]: *** [tools/make/models.mk:103: download-models-full] Error 1

To Reproduce

visit: https://github.com/vllm-project/semantic-router/actions/runs/20147145863/job/57830528131#step:13:231

Expected behavior

No response

Additional context

@Xunzhuo @rootfs

Should we need add HF_TOKEN in github action secrets? or should not use restricted models.

[samzong]

seem part of #816

[srini-abhiram]

Gemma300 is not restricted, you have to log in HF and accept google terms. It should work from then, I faced the same issue.