ComfyUI for Developers: SDKs, Server API, and Cloud

Build on ComfyUI with official Python and TypeScript SDKs, run it as a server, call Cloud APIs, or connect AI agents via MCP.

ComfyUI is a modular GenAI inference engine that can be run as a server, accessed via API, extended with custom nodes, and managed from the command line. Most API work follows two steps: get ComfyUI running somewhere, then run workflows against it from your application.

Use cases

Call a hosted model

Generate with hosted models through Comfy Router.

Deploy ComfyUI

Deploy ComfyUI on the Developer Platform or your own infrastructure.

Run workflows

Run workflows from your application with an SDK or the HTTP API.

Connect AI agents

Connect AI agents to ComfyUI with MCP and Comfy CLI.

Quick Start

The fastest way to try the workflow API is to run a workflow against Comfy Cloud. The SDKs point at Comfy Cloud by default. You’ll need a workflow, an API key, and a paid Comfy Cloud subscription.

1. Get a workflow

   Browse ComfyUI Workflows for community workflows, or build your own in the editor. Save it in API format as workflow_api.json.

2. Install an SDK

   Python

   pip install comfy-sdk
   
3. Run it

   Python

   from comfy_sdk import Comfy
   
   client = Comfy(api_key="comfyui-...")  # Comfy Cloud is the default target
   
   wf = client.workflows.from_file("workflow_api.json")
   
   job = client.run(wf)
   for output in job.get_outputs("9"):
       output.to_file(output.name)
   

   "9" is the ID of the output node in your workflow file. See the SDK guide for inputs, progress events, and error handling.

4. Deploy your own ComfyUI

   When you need your own models and custom nodes behind the endpoint, create a Comfy API deployment, a serverless endpoint managed through the Developer Platform. The same SDK code works against it; only the base URL changes.