HuggingfaceInferenceAPI

You wanna set custom url in text embedding inference other than the default?

You can paste in a URL directly for the model_name

As long as it points to an inference endpoint it will work

@Logan M oh, I didn't know that. Can you share an example for huggingface jinaai model?

@WhiteFang_Jr yeah it's correct

Do you have jinaai deployed on the inference api?

yes

I run it locally

using TEI

oh then you can use the TEI embed model class

embed_model = TextEmbeddingsInference(
model_name="BAAI/bge-large-en-v1.5", # required for formatting inference text,
timeout=60, # timeout in seconds
embed_batch_size=10, # batch size for embedding
base_url="http://127.0.0.1:8080"
)

nice, let me try it