Implementation:Mistralai Client python Chat Complete

Knowledge Sources	Mistral Client Python Mistral AI Docs
Domains	NLP, LLM_Inference
Last Updated	2026-02-15 14:00 GMT

Overview

Concrete tool for sending synchronous and asynchronous chat completion requests provided by the mistralai SDK's Chat resource.

Description

The Chat.complete() and Chat.complete_async() methods send a list of conversation messages to the Mistral API and return a complete ChatCompletionResponse. The methods handle request serialization, HTTP transport, error handling, and response deserialization. They support structured output via response_format, function calling via tools, and safety prompts via safe_prompt.

Usage

Call client.chat.complete() for synchronous usage or client.chat.complete_async() with await for async usage. This is the primary method for non-streaming chat interactions.

Code Reference

Source Location

Repository: client-python
File: src/mistralai/client/chat.py
Lines: L108-283 (sync), L285-460 (async)

Signature

class Chat:
    def complete(
        self,
        *,
        model: str,
        messages: List[ChatCompletionRequestMessage],
        temperature: Optional[float] = None,
        top_p: Optional[float] = None,
        max_tokens: Optional[int] = None,
        response_format: Optional[ResponseFormat] = None,
        tools: Optional[List[Tool]] = None,
        tool_choice: Optional[Union[ToolChoiceEnum, ToolChoice]] = None,
        safe_prompt: Optional[bool] = None,
        random_seed: Optional[int] = None,
        prediction: Optional[Prediction] = None,
        parallel_tool_calls: Optional[bool] = None,
    ) -> ChatCompletionResponse:
        ...

    async def complete_async(
        self,
        *,
        model: str,
        messages: List[ChatCompletionRequestMessage],
        # Same parameters as complete()
    ) -> ChatCompletionResponse:
        ...

Import

from mistralai import Mistral
# Access via: client.chat.complete(...)

I/O Contract

Inputs

Name	Type	Required	Description
model	str	Yes	Model ID (e.g., "mistral-large-latest")
messages	List[ChatCompletionRequestMessage]	Yes	Conversation messages
temperature	Optional[float]	No	Sampling temperature (0.0-0.7)
top_p	Optional[float]	No	Nucleus sampling threshold
max_tokens	Optional[int]	No	Maximum output tokens
response_format	Optional[ResponseFormat]	No	JSON or text output format
tools	Optional[List[Tool]]	No	Tool definitions for function calling
safe_prompt	Optional[bool]	No	Inject safety system prompt

Outputs

Name	Type	Description
response	ChatCompletionResponse	Contains choices, usage info, model metadata
response.choices	List[ChatCompletionChoice]	Generated completions (usually 1)
response.usage	UsageInfo	Token counts (prompt, completion, total)

Usage Examples

Synchronous Chat

import os
from mistralai import Mistral
from mistralai.models import UserMessage

client = Mistral(api_key=os.environ["MISTRAL_API_KEY"])

response = client.chat.complete(
    model="mistral-large-latest",
    messages=[
        UserMessage(content="What is the capital of France?"),
    ],
)

print(response.choices[0].message.content)

Async Chat

import asyncio
import os
from mistralai import Mistral
from mistralai.models import UserMessage

async def main():
    client = Mistral(api_key=os.environ["MISTRAL_API_KEY"])

    response = await client.chat.complete_async(
        model="mistral-large-latest",
        messages=[
            UserMessage(content="Explain quantum computing briefly."),
        ],
    )

    print(response.choices[0].message.content)

asyncio.run(main())

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment