ComfyUI, a startup providing creators with control over image, video, and audio outputs from diffusion models through a node-based workflow, has secured a $30 million funding round at a valuation of $500 million. The round was led by Craft Ventures, with contributions from Pace Capital, Chemistry, and TruArrow.
ComfyUI originated as an open source project in 2023, following the introduction of diffusion models. At that time, models such as Midjourney and OpenAI’s DALL-E faced challenges, often making significant errors like adding extra fingers to hands. To overcome these issues, the project founders created a modular framework offering creators detailed control over each step of the generation process.
The tool gained considerable popularity among creative professionals, leading to its evolution into a formal startup. In late 2024, ComfyUI raised $19 million in a Series A financing from investors like Chemistry Ventures, Cursor Capital, and Guillermo Rauch, founder of Vercel.
While newer diffusion models have improved, the demand for the detailed precision ComfyUI provides has increased. ComfyUI’s node-based interface allows creators to directly manage specific components of the generation process, ensuring high-quality final outputs.
The tool has amassed over 4 million users and is utilized by creative professionals in visual effects, animation, advertising, and industrial design. The startup notes that “ComfyUI artist or engineer” is now a common job title on studio job boards.
Despite advancements in video and image models, Yan believes these tools remain imperfect, maintaining that ComfyUI will continue to be in high demand. ComfyUI faces competition from Weavy, a startup acquired by Figma last year.