Groq MCP
AIstdio
Groq's official MCP server, giving agents access to Groq's high-speed inference stack. Supports fast chat and reasoning completions, vision models for interpreting images, text-to-speech and speech-to-text, batch processing for high-volume workloads, and lookups against Groq's own documentation.
Connect
Add this configuration to .claude/mcp.json
{
"mcpServers": {
"groq": {
"command": "npx",
"args": [],
"env": {
"GROQ_API_KEY": "<YOUR_GROQ_API_KEY>"
}
}
}
}Tools (3)
chat_completion
Run a chat completion against a Groq-hosted model.
{
"type": "object",
"required": [
"model",
"prompt"
],
"properties": {
"model": {
"type": "string",
"description": "Groq model identifier"
},
"prompt": {
"type": "string",
"description": "Prompt text"
}
}
}transcribe_audio
Transcribe an audio file to text using a Groq speech model.
{
"type": "object",
"required": [
"file_path"
],
"properties": {
"file_path": {
"type": "string",
"description": "Path to the audio file"
}
}
}text_to_speech
Generate speech audio from text.
{
"type": "object",
"required": [
"text"
],
"properties": {
"text": {
"type": "string",
"description": "Text to synthesize"
},
"voice": {
"type": "string",
"description": "Voice identifier"
}
}
}Resources
This server does not expose any resources.
Prompts
This server does not expose any prompts.
Server information
- Author
- Groq (@groq)
- Repository
- https://github.com/groq/groq-mcp-server
- License
- MIT
- Stars
- 0
- Last updated
- August 3, 2026