Do all the models we upload to Hugging Face need to be deployed in order to be usable via API? I want to make sure because some models I’ve uploaded aren’t being recognized in n8n, and I suspect it’s because they haven’t been deployed. Is there a way to make a model accessible via API without deploying it, or is deployment always required?
Hey @nina1
!
Using HTTP node is not required to be deployed since HF offers Serverless API.
About your
I think HF is a compatibility issue.
Should the URL you use for the API start with router.huggingface?
I entered the Hugging Face URL in the HTTP Request node.
Actually, I uploaded an image classification model.
The url should point to your model eg:
And in the HTTP node you need to add as well the header content type since is an image.
1 Like
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
