Mistral Plugin

<Warning> This plugin is not yet available in the Eliza plugin registry. To use Mistral models today, configure them through the [OpenRouter plugin](/plugin-registry/llm/openrouter) using the appropriate model ID. </Warning>

The Mistral plugin connects Eliza agents to Mistral AI models, offering both open-weight and commercial models with competitive performance, especially for European languages and code generation.

On-demand plugin. This plugin is resolved from the remote elizaOS plugin registry and auto-installs when its API key is detected. It is not included in Eliza's bundled plugins.json index.

Package: @elizaos/plugin-mistral

Installation

bash

eliza plugins install @elizaos/plugin-mistral

Auto-Enable

The plugin auto-enables when MISTRAL_API_KEY is present:

bash

export MISTRAL_API_KEY=your-mistral-api-key

Configuration

Environment Variable	Required	Description
`MISTRAL_API_KEY`	Yes	Mistral AI API key from console.mistral.ai

eliza.json Example

json

{
  "auth": {
    "profiles": {
      "default": {
        "provider": "mistral",
        "model": "mistral-large-latest"
      }
    }
  }
}

Supported Models

Model	Context	Best For
`mistral-large-latest`	128k	Complex reasoning, multilingual
`mistral-medium-latest`	32k	Balanced performance
`mistral-small-latest`	32k	Fast, cost-efficient tasks
`open-mistral-nemo`	128k	Long-context, open-weight
`codestral-latest`	32k	Code generation and completion
`mistral-embed`	8k	Text embeddings

Features

Streaming responses
Tool use / function calling
Strong multilingual performance (especially European languages)
Code generation (Codestral)
Embeddings for semantic search
Structured JSON output
Guardrail-friendly moderation endpoint

Groq Plugin — Fast inference for open-weight models
OpenRouter Plugin — Route between Mistral and other providers
Model Providers — Compare all providers