ComfyUI hits $500M valuation as creators seek more control over AI-generated media
Summary (EN)
TechCrunch reported on April 24 that ComfyUI has raised $30 million at a $500 million valuation, marking a significant financing event for one of the most widely used workflow tools in AI-generated media. The company began as an open-source project in 2023, shortly after diffusion models entered mainstream use, and focused on a problem that simple prompt interfaces did not solve: precise, repeatable control over generation pipelines. The report says ComfyUI’s node-based system lets creators define and link individual steps in image, video, and audio generation, which reduces the randomness that often appears when users try to modify outputs through repeated prompting alone. Chief executive Yoland Yan told TechCrunch that mainstream prompt-driven systems may get users only 60% to 80% of the way to a desired result, while the final refinements often feel like pulling a slot machine because small edits can unexpectedly overwrite successful elements. ComfyUI’s pitch is that professional creators need more granular control than chat-style interfaces can reliably provide. According to the article, the product now claims more than 4 million users and is used in visual effects, animation, advertising, and industrial design, to the point that “ComfyUI artist or engineer” is appearing as a job title in studio listings. The financing is notable because it suggests investors see enduring commercial value not only in foundation models, but in the workflow infrastructure that helps professionals turn generative systems into controllable production tools.
Summary (ZH)
TechCrunch 于 4 月 24 日报道称，ComfyUI 完成 3000 万美元融资，估值达到 5 亿美元。这是 AI 生成媒体工作流工具领域一次相当重要的融资事件。ComfyUI 最初在 2023 年作为开源项目出现，时间点正值扩散模型开始进入主流视野。它瞄准的问题并不是“能不能生成内容”，而是简单提示词界面无法解决的另一层需求，即对图像、视频和音频生成流程进行精细、可重复、可控的管理。报道指出，ComfyUI 的节点式工作流允许创作者拆分并连接生成过程中的每一个步骤，从而减少单纯反复改写提示词所带来的随机性。公司 CEO Yoland Yan 向 TechCrunch 表示，主流提示词式工具通常只能把用户带到目标结果的 60% 到 80%，而剩余 20% 的细节修改像拉老虎机，因为一次微调往往会破坏原本已经满意的部分。ComfyUI 的核心主张是，专业创作者需要的不是更花哨的聊天界面，而是更细粒度的创作控制。报道称，该产品现已拥有超过 400 万用户，并被用于视觉特效、动画、广告和工业设计等场景，甚至在部分工作室招聘中开始出现“ComfyUI artist or engineer”这样的职位名称。这笔融资的重要意义在于，它说明投资人不仅看重基础模型本身，也开始持续押注那些能把生成式 AI 变成可控生产工具的工作流基础设施。
Source