ComfyUI, the open-source platform that lets you build custom AI image and video generation pipelines, just raised $30 million. The company announced the round today, which brings its valuation to $500 million.
Generative AI has gotten absurdly good — anyone can type a prompt and get a professional-looking image. But that's also the problem. A single prompt can't reliably reproduce the same result, and it's impossible to manage complex production workflows with one-off generations. Professionals need control and repeatability, and that's where ComfyUI comes in.
ComfyUI's answer is the "workflow" — instead of treating generation as a one-shot operation, you define a series of steps that can be saved, reused, and shared. Once you build a pipeline, you can run it again, tweak it, or hand it off to a teammate.
Under the hood, workflows are made of nodes — modular blocks that handle specific tasks like generating an image, transforming it, or editing it. Connect the nodes together and you've got a complex pipeline that's far more flexible than traditional AI tools.
ComfyUI is open-source, so developers and creators can add their own nodes or extend existing functionality. Millions of people already use it — a sign that the demand for more control and flexibility in AI production is real.
With this $30 million infusion, the company says it will deliver "better infrastructure, better tools, and better support for the ecosystem you've built." And it emphasized that ComfyUI will always stay open.
It's part of a broader push for open-source AI that some lawmakers are trying to curb with anti-distillation legislation.