mirascope.core.litellm.call

See OpenAI For Full Reference

We use the types found in the openai module for our LiteLLM integration, so please reference that module for information regarding anything beyond the LiteLLM call decorator.

A decorator for calling the LiteLLM API with a typed function.

Usage Documentation

Calls

This decorator is used to wrap a typed function that calls the LiteLLM API. It parses the prompt template of the wrapped function as the messages array and templates the input arguments for the function into each message's template.

Example:

from mirascope.core import prompt_template
from mirascope.core.litellm import litellm_call


@litellm_call("gpt-4o-mini")
def recommend_book(genre: str) -> str:
    return f"Recommend a {genre} book"

response = recommend_book("fantasy")
print(response.content)

Parameters:

Name	Type	Description	Default
model	str	The model to use in the API call.	required
stream	bool	Whether to stream the response from the API call.	required
tools	list[BaseTool | Callable]	The tools to use in the API call.	required
response_model	BaseModel | BaseType	The response model into which the response should be structured.	required
output_parser	Callable[[OpenAICallResponse | ResponseModelT], Any]	A function for parsing the call response whose value will be returned in place of the original call response.	required
json_mode	bool	Whether to use JSON Mode.	required
client	None	LiteLLM does not support a custom client.	required
call_params	OpenAICallParams	The OpenAICallParams call parameters to use in the API call.	required

Returns:

Name	Type	Description
decorator	Callable	The decorator for turning a typed function into a LiteLLM routed LLM API call.