Groq

Installation

TypingMindClaude DesktopCursorVS CodeClaude CodeGemini CLICodexWindsurf

TypingMind

Prerequisites:

Node.js 18+

Copy

{
  "mcpServers": {
    "groq": {
      "command": "npx",
      "args": [
        "-y",
        "groq-mcp"
      ],
      "env": {
        "GROQ_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

Available Tools

• list_models

  Retrieve available LLM models from Groq's API including Llama, Mixtral, and other supported open-source models

• create_completion

  Generate text completion using specified Groq model with ultra-fast LPU inference

• create_chat_completion

  Generate conversational responses using chat-formatted input with Groq's high-speed inference

• get_model_details

  Fetch detailed information about a specific model including parameters, context length, and capabilities

• create_streaming_completion

  Generate streaming text completion for real-time response delivery

• create_streaming_chat

  Generate streaming chat responses with token-by-token delivery

• get_usage_stats

  Retrieve API usage statistics including tokens consumed, requests made, and rate limit status

• validate_api_key

  Verify Groq API key validity and check authentication status

• estimate_tokens

  Calculate estimated token count for input text before making API calls

• get_rate_limits

  Fetch current rate limit information and remaining quota for the API key

• list_supported_languages

  Get list of natural languages supported by available models

• get_model_performance

  Retrieve performance metrics including inference speed and throughput for specific models

• create_batch_completion

  Process multiple completion requests in a single batch operation

• get_system_status

  Check Groq service status and LPU infrastructure availability

• configure_inference_params

  Set inference parameters like temperature, max tokens, and top-p for model requests