Can i pass the prompt via a post request to the HF Inference endpoint / API?
#5
by jamesdhope - opened
Hi there, can I pass the prompt in the post request to the HF inferencing endpoint, and what is the correct payload format please? Thanks.
I am not sure if this model is already supported in inferencing endpoint or not (I am not in that team). I know it's not supported yet in transformers' pipeline.
I will check the inferencing endpoint part.
Is there any update on this issue?
Hi~
For this model, at this moment, a custom Inference handler is needed. You can check the following documentation
https://huggingface.co/docs/inference-endpoints/guides/custom_handler