Maximum strings in a batch for serverless feature extraction

Inference Endpoints on the Hub
1

I am using the serverless API to call an embedding model intfloatmultilingual-e5-large. I batch up a number of strings into a list

{ inputs: batch_strings}

Where batch_strings is of type List[int]. What is the maximum number of strings that I can batch up in a call to this API? What about string length?

1 Like