📌Overview

Purpose: The MCP Server is designed to facilitate interaction with fal.ai models and services through the Model Context Protocol.

Overview: This framework provides a comprehensive set of tools to manage, search, and utilize fal.ai models efficiently, allowing developers to integrate advanced AI capabilities into their applications with ease.

Key Features:

Model Management: Lists all available fal.ai models and supports keyword searching, enabling users to quickly find and access the desired models.
Content Generation: Facilitates content generation using selected models, with options for both direct execution and queued processing for better resource management.
Queue Management: Offers status checking, results retrieval, and cancellation of queued requests, streamlining the handling of multiple model executions.
File Upload Capability: Allows users to upload files to fal.ai's CDN, enhancing the versatility of model interactions.

fal.ai MCP Server

A Model Context Protocol (MCP) server for interacting with fal.ai models and services.

Features

List all available fal.ai models
Search for specific models by keywords
Get model schemas
Generate content using any fal.ai model
Support for both direct and queued model execution
Queue management (status checking, getting results, cancelling requests)
File upload to fal.ai CDN

Requirements

Python 3.10+
fastmcp
httpx
aiofiles
A fal.ai API key

Installation

Clone this repository:

git clone https://github.com/am0y/mcp-fal.git
cd mcp-fal

Install the required packages:

pip install fastmcp httpx aiofiles

Set your fal.ai API key as an environment variable:

export FAL_KEY="YOUR_FAL_API_KEY_HERE"

Usage

Running the Server

Run the server in development mode with:

fastmcp dev main.py

This launches the MCP Inspector web interface to test the tools interactively.

Installing in Claude Desktop

To use the server with Claude Desktop:

fastmcp install main.py -e FAL_KEY="YOUR_FAL_API_KEY_HERE"

This makes the server available to Claude in the Desktop app.

Running Directly

You can also run the server directly:

python main.py

API Reference

Tools

models(page=None, total=None) - List available models with optional pagination
search(keywords) - Search for models by keywords
schema(model_id) - Get OpenAPI schema for a specific model
generate(model, parameters, queue=False) - Generate content using a model
result(url) - Get result from a queued request
status(url) - Check status of a queued request
cancel(url) - Cancel a queued request
upload(path) - Upload a file to fal.ai CDN

License

MIT License