calls

Class AsyncCall

A class for generating responses using LLMs asynchronously.

Bases: BaseCall[P, AsyncPrompt, AsyncToolkit, FormattableT], Generic[P, FormattableT]

Function call

Generates a response using the LLM asynchronously.

Function stream

Generates a streaming response using the LLM asynchronously.

Class AsyncContextCall

A class for generating responses using LLMs asynchronously.

Bases: BaseCall[P, AsyncContextPrompt, AsyncContextToolkit[DepsT], FormattableT], Generic[P, DepsT, FormattableT]

Function call

Generates a response using the LLM asynchronously.

Function stream

Generates a streaming response using the LLM asynchronously.

Class Call

A class for generating responses using LLMs.

Bases: BaseCall[P, Prompt, Toolkit, FormattableT], Generic[P, FormattableT]

Function call

Generates a response using the LLM.

Function stream

Generates a streaming response using the LLM.

Class CallDecorator

A decorator for converting prompts to calls.

Bases:

Class ContextCall

A class for generating responses using LLMs.

Bases: BaseCall[P, ContextPrompt, ContextToolkit[DepsT], FormattableT], Generic[P, DepsT, FormattableT]

Function call

Generates a response using the LLM.

Function stream

Generates a streaming response using the LLM.

Function call

Returns a decorator for turning prompt template functions into generations.

This decorator creates a Call or ContextCall that can be used with prompt functions. If the first parameter is typed as llm.Context[T], it creates a ContextCall. Otherwise, it creates a regular Call.

Example:

Regular call:

from mirascope import llm

@llm.call(
    provider="openai:completions",
    model_id="gpt-4o-mini",
)
def answer_question(question: str) -> str:
    return f"Answer this question: {question}"

response: llm.Response = answer_question("What is the capital of France?")
print(response)

Example:

Context call:

from dataclasses import dataclass
from mirascope import llm

@dataclass
class Personality:
    vibe: str

@llm.call(
    provider="openai:completions",
    model_id="gpt-4o-mini",
)
def answer_question(ctx: llm.Context[Personality], question: str) -> str:
    return f"Your vibe is {ctx.deps.vibe}. Answer this question: {question}"

ctx = llm.Context(deps=Personality(vibe="snarky"))
response = answer_question(ctx, "What is the capital of France?")
print(response)