HearmemanAI

You can execute ComfyUI image and video generation on remote RunPod servers, submit jobs to serverless endpoints, check generation status, download models, and configure remote ComfyUI settings. It covers both job execution and model management on rented GPU infrastructure.

It targets AI content creators and engineers who generate images or videos with ComfyUI but lack local GPU hardware. Anyone mentioning ComfyUI pipelines, RunPod image generation, remote GPU rendering, or serverless ComfyUI execution is the intended user.

The comfy-gen command interface connects to a RunPod serverless endpoint hosting ComfyUI. You submit a generation request, the remote GPU server renders it, and you poll for status until completion. Models can be downloaded to the remote server beforehand.

You need access to a RunPod serverless endpoint running ComfyUI and the comfy-gen command interface installed locally. A ComfyUI generation definition is required for each submission, since the skill executes existing ComfyUI pipelines rather than building them.