Hidream I1 Full

You run a local MCP server that exposes the fal-ai/hidream-i1-full image generation model. With this server you can generate high-quality images through the fal.ai platform, control generation parameters, and manage long-running requests via a queue or streaming progress.

How to use

Connect to the server using your MCP client with the local stdio configuration. Start the server locally and reference the included command and arguments to launch it. Once running, you can submit image-generation requests using the available tools, monitor real-time progress for streaming generation, or enqueue long-running jobs and check their status and results. You also have options to apply LoRA weights, customize image sizes, and enable or disable the safety checker.

How to install

Prerequisites: you need Node.js and npm installed on your machine. You also need access to a fal.ai API key.

# 1. Clone the MCP server repository
git clone https://github.com/PierrunoYT/fal-hidream-i1-full-mcp-server.git
cd fal-hidream-i1-full-mcp-server

# 2. Install dependencies
npm install

# 3. Build the project
npm run build

# 4. Set up your API key for fal.ai (example)
export FAL_KEY="your_fal_api_key_here"

Additional configuration and security

Set your fal.ai API key as an environment variable to authenticate requests. The key is required for all image generation operations.

{
  "type": "stdio",
  "name": "hidream_i1_full",
  "command": "node",
  "args": ["/path/to/fal-hidream-i1-full-mcp-server/build/index.js"],
  "env": {
    "FAL_KEY": "your_fal_api_key_here"
  }
}

Usage notes and behavior

• Images are downloaded locally to an images/ directory with descriptive filenames. The response includes the local file paths, the original URLs, image dimensions, content types, generation parameters used, and any tracking IDs.

• You can enable streaming to receive real-time progress, or submit a queued job for longer requests. Webhook notifications are available for completed queued requests.

• The server supports predefined image sizes and custom dimensions, and you can apply LoRA weights to tailor image styles.

Troubleshooting and tips

If you encounter missing API keys, invalid parameters, network issues, rate limits, or generation failures, check the error details returned by the server and verify your API key is correctly set.

For long-running jobs, monitor the queue using the provided status and result endpoints, and verify webhook delivery if you rely on external notifications.

Notes

The MCP server exposes a set of tools to generate images, stream progress, or manage queued requests. Ensure you keep your API key secure and do not expose it in client configurations or public code.

Available tools

hidream_i1_full_generate

Generate images using the standard synchronous method with configurable prompt, size, steps, seed, guidance, safety, and optional LoRA weights.

hidream_i1_full_generate_stream

Generate images with streaming progress updates, allowing real-time viewing of results as they are produced.

hidream_i1_full_generate_queue

Submit a long-running image generation request to a queue, optionally with a webhook for completion notifications.

hidream_i1_full_queue_status

Check the current status and logs of a queued request to monitor progress.

hidream_i1_full_queue_result

Retrieve the result for a completed queued request, including generated image data and metadata.