ComfyUI hits $500M valuation as creators seek control over AI media

ComfyUI, a startup that enables creators to precisely control image, video, and audio outputs from diffusion models using a node-based workflow, has raised $30 million in funding at a $500 million valuation.
The funding round was led by Craft Ventures, with additional participation from Pace Capital, Chemistry, and TruArrow.
ComfyUI began as an open-source project in 2023, shortly after diffusion models emerged. At the time, models like Midjourney and OpenAI’s DALL-E were still quite unreliable, often producing obvious errors like extra fingers on hands.
To address these limitations, the project founders built a modular framework that gives creators fine-grained control over every step of the generation process.
The tool gained significant traction among creative professionals, eventually leading to the formation of a proper startup. In late 2024, ComfyUI raised $19 million in Series A funding from investors including Chemistry Ventures, Cursor Capital, and Guillermo Rauch (founder of Vercel).
While the latest diffusion models have improved significantly since the days of adding extra fingers, the demand for the granular precision ComfyUI provides has only grown.
"If you consider a typical prompt-based tool like Midjourney or ChatGPT, you ask for something and it gets only 60% to 80% of the way there," said Yoland Yan, co-founder and CEO of ComfyUI. "But to adjust that remaining 20%, you have to keep pulling the slot machine."
Yan likened the process to gambling in a casino, since prompting the model for a small change can lead to a completely different output, potentially overwriting parts that were already perfect.
ComfyUI's node-based interface lets creators connect specific components of the generation workflow, granting them full control over the quality of the final output.
"That level of nuance is hard to communicate in the prompt box of a base model," Yan said.
Creators appear to agree, as ComfyUI reports having over four million users.
The tool is used by creative professionals in visual effects, animation, advertising, and even industrial design.
According to the startup, its platform has become an essential tool for technical artists and other creatives, so much so that job postings for "ComfyUI artist" or "ComfyUI engineer" are now common on studio job boards.
Even as video and image foundation models keep improving, Yan argues they are still far from perfect, ensuring continued high demand for a tool like ComfyUI.
"In a world where AI slop is everywhere, the Comfy version of the human-in-the-loop approach will ultimately capture the most attention," he said.
ComfyUI’s competitors include Weavy, a startup that Figma acquired last year.
Related article
Google Vids Introduces Feature Allowing Users to Star in Custom AI-Generated Videos
Although OpenAI’s Sora has been discontinued, Google believes there remains strong demand for a tool that enables users to appear as the main character in their own AI-generated videos. On Thursday, the company released an update to Google Vids that
Meta Unveils New AI Creator Assistant Integrated into Facebook Platform
On Thursday, Meta revealed the launch of a new AI creator assistant on Facebook, designed to provide creators with personalized suggestions tailored to their content style, performance metrics, community engagement, and strategic goals.Currently, cre
Meta launches subscriptions for Instagram, Facebook, WhatsApp, with AI plans to follow
Meta is intensifying its focus on subscription services. On Wednesday, the social media giant announced the global rollout of consumer subscription plans for its flagship apps — Instagram, Facebook, and WhatsApp — and the start of testing new subscri
Related Special Topic Recommendations
Comments (1)
0/500

ComfyUI, a startup that enables creators to precisely control image, video, and audio outputs from diffusion models using a node-based workflow, has raised $30 million in funding at a $500 million valuation.
The funding round was led by Craft Ventures, with additional participation from Pace Capital, Chemistry, and TruArrow.
ComfyUI began as an open-source project in 2023, shortly after diffusion models emerged. At the time, models like Midjourney and OpenAI’s DALL-E were still quite unreliable, often producing obvious errors like extra fingers on hands.
To address these limitations, the project founders built a modular framework that gives creators fine-grained control over every step of the generation process.
The tool gained significant traction among creative professionals, eventually leading to the formation of a proper startup. In late 2024, ComfyUI raised $19 million in Series A funding from investors including Chemistry Ventures, Cursor Capital, and Guillermo Rauch (founder of Vercel).
While the latest diffusion models have improved significantly since the days of adding extra fingers, the demand for the granular precision ComfyUI provides has only grown.
"If you consider a typical prompt-based tool like Midjourney or ChatGPT, you ask for something and it gets only 60% to 80% of the way there," said Yoland Yan, co-founder and CEO of ComfyUI. "But to adjust that remaining 20%, you have to keep pulling the slot machine."
Yan likened the process to gambling in a casino, since prompting the model for a small change can lead to a completely different output, potentially overwriting parts that were already perfect.
ComfyUI's node-based interface lets creators connect specific components of the generation workflow, granting them full control over the quality of the final output.
"That level of nuance is hard to communicate in the prompt box of a base model," Yan said.
Creators appear to agree, as ComfyUI reports having over four million users.
The tool is used by creative professionals in visual effects, animation, advertising, and even industrial design.
According to the startup, its platform has become an essential tool for technical artists and other creatives, so much so that job postings for "ComfyUI artist" or "ComfyUI engineer" are now common on studio job boards.
Even as video and image foundation models keep improving, Yan argues they are still far from perfect, ensuring continued high demand for a tool like ComfyUI.
"In a world where AI slop is everywhere, the Comfy version of the human-in-the-loop approach will ultimately capture the most attention," he said.
ComfyUI’s competitors include Weavy, a startup that Figma acquired last year.
Google Vids Introduces Feature Allowing Users to Star in Custom AI-Generated Videos
Although OpenAI’s Sora has been discontinued, Google believes there remains strong demand for a tool that enables users to appear as the main character in their own AI-generated videos. On Thursday, the company released an update to Google Vids that
Meta Unveils New AI Creator Assistant Integrated into Facebook Platform
On Thursday, Meta revealed the launch of a new AI creator assistant on Facebook, designed to provide creators with personalized suggestions tailored to their content style, performance metrics, community engagement, and strategic goals.Currently, cre
Meta launches subscriptions for Instagram, Facebook, WhatsApp, with AI plans to follow
Meta is intensifying its focus on subscription services. On Wednesday, the social media giant announced the global rollout of consumer subscription plans for its flagship apps — Instagram, Facebook, and WhatsApp — and the start of testing new subscri





Home






