Image Generation

MCP server enabling high-quality image generation via Together AI's Flux.1 Schnell model.

javascript

9

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "manascb1344-together-mcp-server": {
      "command": "npx",
      "args": [
        "together-mcp@latest -y"
      ],
      "env": {
        "TOGETHER_API_KEY": "<API KEY>"
      }
    }
  }
}

You can generate high-quality images using the Flux.1 Schnell model through an MCP server that exposes a simple, standardized image generation interface. This server lets you tune dimensions, steps, and other parameters while handling errors gracefully, and can optionally save generated images to disk as PNG files.

How to use

You interact with the image generation MCP server by sending a generate_image request to the server. Provide a prompt describing the image you want, and optionally adjust the model, output size, number of steps, image count, and how you want the result returned. You can also save the generated image to disk if you want a local PNG file. Use the defaults if you don’t specify a parameter and make sure any provided values stay within supported ranges.

How to install

Prerequisites: Install Node.js version 16 or newer on your system.

Install the MCP server package and run it locally.

npm install together-mcp

npx together-mcp@latest

Configuration

Configure a local MCP channel for image generation by specifying the MCP server run command and API key in your MCP configuration. The following example shows how to register the image generation server with the MCP runtime.

{
  "mcpServers": {
    "together-image-gen": {
      "command": "npx",
      "args": ["together-mcp@latest -y"],
      "env": {
        "TOGETHER_API_KEY": "<API KEY>"
      }
    }
  }
}

Usage details

The server exposes a tool named generate_image. You provide a prompt as the required parameter and can override defaults for model, width, height, steps, number of images, response format, and an optional path to save the image.

Default values and notes

Default model: black-forest-labs/FLUX.1-schnell-Free. Default width: 1024. Default height: 768. Default steps: 1. Default n: 1. Default response_format: b64_json. All optional parameters are used only if provided; otherwise, sensible defaults apply.

Notes: If you provide image_path, ensure the directory exists and is writable. Base64 responses can be large; use URL format for large images.

Available tools

generate_image

Generates an image based on a text prompt. Optional parameters let you set the model, output width/height, number of steps, number of images, and how the result is returned (base64 data or a URL).