ComfyUI, a startup that helps creators control image, video, and audio output from viral models using node-based workflows, has raised a $30 million funding round at a $500 million valuation.
The round was led by Craft Ventures, with participation from other investors including Pace Capital, Chemistry, and TruArrow.
ComfyUI was started as an open source project in 2023, shortly after the introduction of the diffusion model. At the time, models like Midjourney and OpenAI's DALL-E were barely functional and often made big mistakes like adding extra fingers to the hand.
To address these limitations, the project founders developed a modular framework that gives authors fine-grained control over each step of the generation process.
Their tools gained so much traction among creative professionals that they eventually evolved into an official startup. In late 2024, ComfyUI raised $19 million in Series A funding from investors including Chemistry Ventures, Cursor Capital, and Vercel founder Guillermo Rauch.
Modern popular models have come a long way since adding six digits to your hand, but the need for the granular precision that ComfyUI provides will only continue to grow.
“When you think about typical prompt-based solutions like Midjourney or ChatGPT, they ask for something. [gets only] “To change the other 20%, you need to try this slot machine,” Yolando Jan, co-founder and CEO of ComfyUI, told TechCrunch.
Yang (pictured left) likened the process to playing in a casino, since making small changes to the model can result in completely different outputs, overwriting parts that were already perfect.
ComfyUI's node-based interface allows authors to link specific components of the generation process, giving them complete control over the quality of the final output.
“You can’t just convey that message in a prompt box. [of a foundational model]” said Yang.
It seems the creators agree, as ComfyUI claims to have over 4 million users.
This tool is used by creative professionals in visual effects, animation, advertising, and even industrial design.
The company says its service has become such a necessary tool of the trade for technical artists and other creators that it's not uncommon to see “ComfyUI Artist or Engineer” listed as a job title on studio job boards.
While the basic models for video and images continue to improve, they are far from perfect, and Yan argues that tools like ComfyUI will continue to be in high demand.
“In a world where AI slop is everywhere, Comfy’s version of the human-involved approach will eventually get most of the attention,” he said.
ComfyUI's competitors include startup Weavy, which was acquired by Figma last year.
If you buy through links in our articles, we may earn a small commission. This does not affect editorial independence.