Mistral AI Models

A drag-and-drop component for integrating Mistral AI models into your workflow. Configure model parameters and connect inputs/outputs to other components.

Mistral AI component interface and configuration

API Key Notice: A valid Mistral AI API key is required to use this component. Ensure your API key has sufficient quota and appropriate rate limits for your application needs.

Component Inputs

Input: Text input for the model
Example: "Write a summary of deep learning techniques for computer vision."
System Message: System prompt to guide model behavior
Example: "You are a technical expert specializing in machine learning concepts."
Stream: Toggle for streaming responses
Example: true (for real-time token streaming) or false (for complete response)
Model Name: The Mistral AI model to use
Example: "codestral-latest", "mistral-medium", "mistral-small-latest"
Mistral API Base: API endpoint URL
Example: "https://api.mistral.ai/v1"
Mistral API Key: Your API authentication key
Example: "m-xzy123abcdef789..."

Component Outputs

Text: Generated text output
Example: "Deep learning techniques for computer vision include convolutional neural networks (CNNs), which..."
Language Model: Model information and metadata
Example: model: mistral-medium, usage: {prompt_tokens: 45, completion_tokens: 150, total_tokens: 195}

Model Parameters

Max Tokens

Maximum number of tokens to generate

Default: Model-dependent
Range: 1 to model maximum
Recommendation: Set based on expected response length

Temperature

Controls randomness in the output - higher values increase creativity

Default: 0.5
Range: 0.0 to 1.0
Recommendation: Lower (0.1-0.3) for factual responses, Higher (0.7-0.9) for creative tasks

Top P

Nucleus sampling parameter - controls randomness along with temperature

Default: 1.0
Range: 0.0 to 1.0
Recommendation: Lower values (e.g., 0.9) for more focused text generation

Random Seed

For reproducible outputs across multiple runs

Default: 1
Range: Integer values
Recommendation: Set a specific seed when reproducibility is important

Safe Mode

Controls content filtering for safer outputs

Options: true/false
Recommendation: Enable for public-facing applications

API Configuration

Max Retries

Number of retry attempts for failed requests

Default: 5
Range: 0 to any reasonable number
Recommendation: Increase for critical applications

Timeout

Request timeout in seconds

Default: 60
Range: Any positive number
Recommendation: Increase for longer generations, decrease for time-sensitive applications

Max Concurrent Requests

Limit on concurrent API calls

Default: 3
Range: 1 to any reasonable number
Recommendation: Adjust based on Mistral AI rate limits and your application's needs

Implementation Example

// Basic configuration
const mistralAI = {
  modelName: "mistral-medium",
  mistralApiKey: process.env.MISTRAL_API_KEY,
  systemMessage: "You are a helpful assistant."
};

// Advanced configuration
const advancedMistralAI = {
  modelName: "codestral-latest",
  mistralApiKey: process.env.MISTRAL_API_KEY,
  mistralApiBase: "https://api.mistral.ai/v1",
  maxTokens: 2000,
  temperature: 0.7,
  topP: 0.95,
  randomSeed: 42,
  safeMode: true,
  maxRetries: 8,
  timeout: 120,
  maxConcurrentRequests: 5,
  stream: true
};

// Usage example
async function generateCode(input) {
  const response = await mistralComponent.generate({
    input: input,
    systemMessage: "You are an expert programmer. Write clean, well-documented code.",
    modelName: "codestral-latest",
    temperature: 0.2
  });
  
  return response.text;
}

Use Cases

Code Generation: Use Codestral models for programming assistance and code completion
Content Creation: Generate articles, blog posts, and creative writing with medium or large models
Conversational Agents: Build chatbots and virtual assistants with context awareness
Text Summarization: Condense long documents into concise summaries
Knowledge-Based Applications: Create applications that require access to general knowledge

Useful Resources

Best Practices

Secure API keys using environment variables
Monitor rate limits and concurrent requests
Start with default temperature (0.5) and adjust based on needs
Use system messages for consistent outputs
Enable streaming for real-time responses in interactive applications
Set appropriate timeout values based on expected generation length

Documentation