Groq Setup

Configure Groq LLM for ultra-fast AI inference.

Overview

Groq provides the LLM infrastructure for ClawChan, offering sub-second response times through their LPU (Language Processing Unit) technology. This guide covers setup and configuration.

Getting Your API Key

• Visit console.groq.com and create an account
• Navigate to the API Keys section
• Generate a new API key
• Copy and store the key securely
• Never share or commit your API key

Configuration

Add your Groq API key to the environment:

• Open your .env file in the project root
• Add GROQ_API_KEY with your key value
• Save the file and restart the development server

Available Models

Groq supports several models for different use cases:

• llama-3.1-70b-versatile - Recommended for general use, excellent reasoning
• llama-3.1-8b-instant - Faster responses for simple tasks
• mixtral-8x7b-32768 - Alternative with longer context window

Performance Optimization

• Use streaming for real-time response display
• Cache common prompts to reduce latency
• Manage context window size for efficiency
• Batch requests when possible

Rate Limits

• Free tier: 30 requests per minute, 14,400 per day
• Monitor usage in the Groq console
• Implement retry logic for rate limit errors
• Contact Groq for higher limits if needed

Troubleshooting

• Invalid API key - Verify the key is correct and active
• Rate limit errors - Reduce request frequency
• Timeout errors - Check network connection
• Model unavailable - Try an alternative model