Open brenkao opened 4 months ago
The main thing on my mind for this feature is how this will interact with our prompt_template
parsing. Will users want direct access to the underlying model prompt structure?
e.g. https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-3/
Likely worth looking into how we can propagate the logits of the output as well
Is your feature request related to a problem? Please describe. While HuggingFace is already supported through the
OpenAICall
class (use base_url, model, api_key="-"), we should support HuggingFace InferenceClient through aHuggingFaceCall
class.Describe the solution you'd like Create a
HugginFaceCall
that extendsBaseCall
that follows same principles as other Calls. This means that call params are properly typed in a Pydantic model, responses extendBaseCallResponse
, etc.Additional context See OpenAICall for reference. https://github.com/Mirascope/mirascope/blob/dev/mirascope/openai/calls.py