Sample code using HF
#29
by
vanshils - opened
Would it be possible to provide a sample code for inference of chat completion request using AutoModelForCausalLM and AutoTokenizer which gives same behaviour as mistral-chat?
Thanks for updating the model card @ybelkada .
If possible could you please update the example/or provide a new one with usage of chat template. Currently the template is a little bit hard to find as we have to dive in mistral-common codebase to see how they perform encode_chat_completion_request.
Once again thanks for the example.