Calling Inference API for image embedding

I’ve been playing around with the Inference API and got the hang of using{model_id} for text data.

Now, I’m curious if there’s a way to do something similar for images. I’m all about getting those image embeddings using the Inference API. Is there a direct way to make this happen? If not, do you suggest any tricks or other methods to get image embeddings through the Inference API?