Mistral AI LLM plugin guide | LiveKit Documentation

Available in

Python

Overview

This plugin allows you to use Mistral AI as an LLM provider for your voice agents.

Installation

Install the LiveKit Mistral AI plugin from PyPI:

uv add "livekit-agents[mistralai]~=1.5"

Authentication

The Mistral AI integration requires a Mistral AI API key.

Set the MISTRAL_API_KEY in your .env file.

Usage

Use Mistral AI within an AgentSession or as a standalone LLM service. For example, you can use this LLM in the Voice AI quickstart.

from livekit.plugins import mistralai

session = AgentSession(
    llm=mistralai.LLM(
        model="mistral-medium-latest"
    ),
    # ... tts, stt, vad, turn_handling, etc.
)

Parameters

This section describes some of the available parameters. See the plugin reference for a complete list of all available parameters.

modelstring | ChatModelsDefault: ministral-8b-latest

Which Mistral AI model to use. You can pass a string or a typed enum from ChatModels.

temperaturefloat

Sampling temperature that controls the randomness of the model's output. Higher values make the output more random, while lower values make it more focused and deterministic. Range of valid values can vary by model.

top_pfloat

Nucleus sampling parameter.

presence_penaltyfloat

Penalize new tokens based on their presence in the text so far.

frequency_penaltyfloat

Penalize new tokens based on their frequency in the text so far.

random_seedint

Random seed for reproducibility.

tool_choiceToolChoice | Literal['auto', 'required', 'none']

Controls how the model uses tools. String options are as follows:

'auto': Let the model decide.
'required': Force tool usage.
'none': Disable tool usage.

max_completion_tokensfloat

The maximum number of tokens the LLM can output.

Provider tools

The Mistral Conversations API supports the following provider tools that enable the model to use built-in capabilities executed on the model server. These tools can be used alongside function tools defined in your agent's codebase.

Tool	Description	Parameters
`WebSearch`	Search the internet for up-to-date information.	None
`DocumentLibrary`	Search uploaded document collections via Libraries.	`library_ids` (required)
`CodeInterpreter`	Write and Execute Python code in a sandboxed environment.	None

from livekit.plugins import mistralai

agent = MyAgent(
    llm=mistralai.LLM(model="mistral-medium-latest"),
    tools=[mistralai.tools.WebSearch()],  # replace with any supported provider tool
)

Additional resources

The following resources provide more information about using Mistral AI with LiveKit Agents.

Python package

The livekit-plugins-mistralai package on PyPI.

Plugin reference

Reference for the Mistral AI LLM plugin.

GitHub repo

View the source or contribute to the LiveKit Mistral AI LLM plugin.

Mistral AI STT docs

Mistral AI STT documentation.

Mistral AI docs

Mistral AI platform documentation.

Voice AI quickstart

Get started with LiveKit Agents and Mistral AI.