Mistral Provider

Overview

The Mistral provider gives you access to Mistral AI’s models for chat completion and embeddings, optimized for European languages and multilingual tasks.

Installation

npm install @core-ai/mistral

createMistral()

Create a Mistral provider instance.

import { createMistral } from '@core-ai/mistral';

const mistral = createMistral({
  apiKey: process.env.MISTRAL_API_KEY
});

Options

apiKey

string

Your Mistral API key. Defaults to MISTRAL_API_KEY environment variable.

baseURL

string

Custom base URL for API requests. Useful for proxies or self-hosted deployments.

client

Mistral

Provide your own configured Mistral client instance.

Returns

MistralProvider - Provider instance with methods to create models.

Provider Methods

chatModel()

Create a chat model instance.

const model = mistral.chatModel('mistral-large-2');

modelId

string

required

Model identifier. See Supported Models below.

embeddingModel()

Create an embedding model instance.

const embeddings = mistral.embeddingModel('mistral-embed');

modelId

string

required

Embedding model identifier.

Supported Models

Chat Models

Large Models

Flagship models for complex tasks.

mistral-large-2 - Latest flagship model
mistral-large - Previous generation flagship

Medium Models

Balanced performance and efficiency.

mistral-medium - Strong performance at lower cost

Small Models

Fast and efficient for simpler tasks.

mistral-small - Quick responses
mistral-tiny - Ultra-fast, lightweight

Specialized Models

Purpose-built for specific use cases.

codestral - Optimized for code generation
mixtral-8x7b - Mixture of experts architecture
mixtral-8x22b - Larger mixture of experts

Embedding Models

mistral-embed - High-quality text embeddings

Capabilities

Feature	Support
Chat Completion	✓
Streaming	✓
Function Calling	✓
Vision	Limited
Reasoning Effort	✗
Embeddings	✓
Image Generation	✗

Mistral models do not support explicit reasoning effort control like OpenAI or Anthropic.

Examples

Basic Chat

import { createMistral } from '@core-ai/mistral';
import { generateText } from '@core-ai/core-ai';

const mistral = createMistral({
  apiKey: process.env.MISTRAL_API_KEY
});

const result = await generateText({
  model: mistral.chatModel('mistral-large-2'),
  prompt: 'Explain the concept of recursion'
});

console.log(result.text);

Multilingual Chat

// Mistral models excel at multilingual tasks
const result = await generateText({
  model: mistral.chatModel('mistral-large-2'),
  prompt: 'Expliquez la théorie de la relativité en français'
});

console.log(result.text);

Streaming

import { streamText } from '@core-ai/core-ai';

const stream = await streamText({
  model: mistral.chatModel('mistral-medium'),
  prompt: 'Write a short story about artificial intelligence'
});

for await (const chunk of stream) {
  if (chunk.type === 'text') {
    process.stdout.write(chunk.text);
  }
}

Code Generation

// Use Codestral for code tasks
const result = await generateText({
  model: mistral.chatModel('codestral'),
  prompt: `
    Write a Python function to calculate the Fibonacci sequence
    up to n terms using dynamic programming.
  `
});

console.log(result.text);

Function Calling

import { generateText, tool } from '@core-ai/core-ai';
import { z } from 'zod';

const result = await generateText({
  model: mistral.chatModel('mistral-large-2'),
  prompt: 'What is 25 multiplied by 4?',
  tools: {
    calculator: tool({
      description: 'Perform mathematical calculations',
      parameters: z.object({
        operation: z.enum(['add', 'subtract', 'multiply', 'divide']),
        a: z.number(),
        b: z.number()
      }),
      execute: async ({ operation, a, b }) => {
        const ops = {
          add: a + b,
          subtract: a - b,
          multiply: a * b,
          divide: a / b
        };
        return { result: ops[operation] };
      }
    })
  }
});

Embeddings

import { embed } from '@core-ai/core-ai';

const result = await embed({
  model: mistral.embeddingModel('mistral-embed'),
  value: 'Search query for semantic similarity'
});

console.log(result.embedding); // number[]

Batch Embeddings

import { embedMany } from '@core-ai/core-ai';

const documents = [
  'First document',
  'Second document',
  'Third document'
];

const results = await embedMany({
  model: mistral.embeddingModel('mistral-embed'),
  values: documents
});

for (const result of results) {
  console.log(result.embedding);
}

Custom Base URL

// Use with self-hosted or proxy endpoints
const mistral = createMistral({
  apiKey: 'your-api-key',
  baseURL: 'https://your-proxy.com/v1'
});

Conversation

const result = await generateText({
  model: mistral.chatModel('mistral-large-2'),
  messages: [
    {
      role: 'user',
      content: 'What is machine learning?'
    },
    {
      role: 'assistant',
      content: 'Machine learning is a subset of AI...'
    },
    {
      role: 'user',
      content: 'Can you give me an example?'
    }
  ]
});

Error Handling

import { APIError } from '@core-ai/core-ai';

try {
  const result = await generateText({
    model: mistral.chatModel('mistral-large-2'),
    prompt: 'Hello!'
  });
} catch (error) {
  if (error instanceof APIError) {
    console.error('Mistral API error:', error.message);
    console.error('Status:', error.statusCode);
  }
}

Best Practices

Model Selection

mistral-large-2 - Complex reasoning, analysis, creative writing
mistral-medium - General purpose, good balance
mistral-small - Simple queries, high throughput
codestral - Code generation, technical documentation
mixtral-8x22b - When you need the best quality

Multilingual Usage

Mistral models have strong support for European languages
Particularly good for French, German, Spanish, Italian
Works well for code-switching between languages

Performance Optimization

Use smaller models for simple tasks to reduce latency and cost
Enable streaming for long-form content
Batch embeddings when processing multiple documents

Model Comparison

Model	Parameters	Best For	Speed
mistral-large-2	Large	Complex reasoning	Slower
mistral-medium	Medium	General purpose	Medium
mistral-small	Small	Simple queries	Fast
codestral	Specialized	Code generation	Medium
mixtral-8x22b	141B (MoE)	Highest quality	Slower
mixtral-8x7b	46.7B (MoE)	Good balance	Medium

Use Cases

Code Generation

Use Codestral for implementing functions, debugging, and technical documentation.

Multilingual Support

Leverage strong European language support for international applications.

Semantic Search

Use embeddings for document search and similarity matching.

Content Generation

Generate articles, summaries, and creative content with Large models.

OpenAI Provider

GPT models with reasoning effort control

Anthropic Provider

Claude models with extended thinking

Embeddings Guide

Learn how to use embeddings effectively

Core

Providers

Mistral Provider

Overview

Installation

createMistral()

Options

Returns

Provider Methods

chatModel()

embeddingModel()

Supported Models

Chat Models

Embedding Models

Capabilities

Examples

Basic Chat

Multilingual Chat

Streaming

Code Generation

Function Calling

Embeddings

Batch Embeddings

Custom Base URL

Conversation

Error Handling

Best Practices

Model Comparison

Use Cases

Code Generation

Multilingual Support

Semantic Search

Content Generation

OpenAI Provider

Anthropic Provider

Embeddings Guide

Core

Providers

​Overview

​Installation

​createMistral()

​Options

​Returns

​Provider Methods

​chatModel()

​embeddingModel()

​Supported Models

​Chat Models

​Embedding Models

​Capabilities

​Examples

​Basic Chat

​Multilingual Chat

​Streaming

​Code Generation

​Function Calling

​Embeddings

​Batch Embeddings

​Custom Base URL

​Conversation

​Error Handling

​Best Practices

​Model Comparison

​Use Cases

Code Generation

Multilingual Support

Semantic Search

Content Generation

​Related

OpenAI Provider

Anthropic Provider

Embeddings Guide

Overview

Installation

createMistral()

Options

Returns

Provider Methods

chatModel()

embeddingModel()

Supported Models

Chat Models

Embedding Models

Capabilities

Examples

Basic Chat

Multilingual Chat

Streaming

Code Generation

Function Calling

Embeddings

Batch Embeddings

Custom Base URL

Conversation

Error Handling

Best Practices

Model Comparison

Use Cases

Related