Skip to main content

Serverless ComfyUI API

Serverless GPU inference for ComfyUI workflows

Why cmfy.cloud?

Serverless GPU

No infrastructure to manage. Submit your ComfyUI workflows via REST API and receive results. Scale from zero to thousands of requests effortlessly.

Smart Caching

Intelligent cache-aware routing minimizes model loading time. Jobs are automatically routed to nodes with pre-loaded models for faster inference.

Fair Queuing

Built-in fair queuing ensures all users get responsive service. Rate limiting and queue depth controls prevent any single user from monopolizing resources.