fal

This MCP server provides an AI video generation service that creates videos from text prompts and/or starting images using two AI models: Luma Ray2 Flash and Kling v1.6 Pro. Users can generate videos from prompts, optionally supply a starting image and/or end image, and control video parameters such as aspect ratio, resolution, duration, and looping. The server exposes two main tools: generate-video for producing new videos and check-video-status for polling the status of a video generation request. When using Claude Desktop or other MCP clients, you can configure the server to point to the local or hosted build, passing necessary environment variables like the FAL API key. The server supports selecting the model (luma or kling) to tailor output style and quality.

Prerequisites:

• Node.js and npm installed on your system
• Git (optional, for cloning repository)

Installation steps:

1. Clone the repository (or download the source): git clone https://github.com/your-org/fal-mcp-server.git cd fal-mcp-server

2. Install dependencies: npm install

3. Create and configure environment variables. Create a .env file or set environment variables in your runtime environment. Required: FAL_KEY=your_fal_key_here You can obtain the API key from FAL.AI.

4. Build the server: npm run build

5. Run the server: npm start

Notes:

• The server entry point after build is typically build/index.js.
• Ensure your environment has network access to FAL.AI services and sufficient credits for video generation.

Tips and considerations:

• Ensure FAL_KEY is valid and has sufficient credits for desired video resolutions and durations.
• High resolutions and longer videos consume more credits and may take longer to generate.
• The two available models offer different strengths: Luma Ray2 Flash for smooth, realistic motion and Kling v1.6 Pro for rich textures and stylized results. Choose based on your prompt and desired outcome.
• In Claude Desktop integrations, configure mcpServers with the appropriate command and args, and pass FAL_KEY via env as shown in the example.
• If you encounter API or generation errors, verify API key validity, credits, prompt correctness, and consider reducing resolution or duration for testing.
• If you update the server code, rebuild before restarting the server to ensure latest changes take effect.