Runway ML, one of the two startups behind the popular AI text-to-image model Stable Diffusion, has raised new funding at a $500 million valuation, Forbes has learned.
Felicis Ventures is leading the new funding, the sources said, which comes on the heels of a boom in generative AI that has captured the public’s attention in recent months thanks to releases that also include OpenAI’s Dall-E and ChatGPT. Runway quickly emerged as one of the buzziest startups in the mix with its video editing software, for which the company has been releasing a bevy of generative AI features. For example, from a photo of a forest, a user can type a short text phrase into Runway’s software and instantly conjure a lake or a castle among the trees.
Sources say annualized revenue hovers around $1 million, the latest indication that while frothy valuations have dissipated from most of the VC industry, the buzzy generative AI space has been an exception. Reached for comment, Runway CEO Cristóbal Valenzuela told Forbes that the company had raised $50 million from Felicis with participation from another new backer Madrona, and existing investors Amplify Partners, Compound VC, Coatue and Lux Capital. He declined to comment on valuation but noted that ARR is now “under $5 million.”
The New York City company was started by Valenzuela, Alejandro Matamala-Ortiz and Anastasis Germanidis. The trio met in art school at New York University, where they bonded over a mutual interest in using digital tools for design. They collaborated on machine learning research that soon evolved into Runway, which was founded in 2018.
The company previously raised a total of $44 million and after a Series B funding round last December, it was valued at $200 million. The company employs 35 people, almost half of whom are immigrants on visas, a choice important to the founders, who are immigrants themselves (Valenzuela and Matamala-Ortiz hail from Chile, Germanidis from Greece). Valenzuela said he would use the new money to bolster research efforts and accelerate its product roadmap. The company will make new hires, such as for senior researchers and for an in-house design team, but had no set headcount target, he said.
“Picture you’re watching a movie and you are the lead actor in the movie: your voice, your body, your face.”
Many Runway customers are individual creatives, who pay $12 per month to use the software. It has also been used by enterprise customers like CBS’s Late Show with Stephen Colbert and the visual effects team for Hollywood hit Everything Everywhere All at Once for video editing. New Balance has used the software to help design athletic shoes. Since the launch of Stable Diffusion, adoption of the product has been “way more mainstream,” according to Valenzuela. In the long run, he envisions building an Adobe-esque suite of AI-native video editing software tools with fantastical applications.
“You are going to be in every movie,” he said in an interview in November. “Picture you’re watching a movie and you are the lead actor in the movie: your voice, your body, your face.”
Valenzuela told Forbes in early November that the company has been able to move fast in augmenting its product because of the head start it accumulated in generative AI long before it became flashy this summer. “It’s not that we shipped [features] so fast. There’s no recipe. It’s that we’ve been training for four years for this marathon,” he said.
“We do research on foundational AI models, we build the infrastructure to use those models and then we build applications,” Valenzuela said. In the case of popular open source text-to-image AI model Stable Diffusion, Runway’s principal research scientist Patrick Esser coauthored the initial research paper alongside academics at the Ludwig Maximilian University of Munich. Runway then helped build the original version of Stable Diffusion and is now implementing the tech into features for its design software. “Stable Diffusion didn’t exist until we invented it,” Valenzuela said in November. Since its public launch in September, the model has been used by more than 10 million people.
Stability AI, another startup commonly associated with the Stable Diffusion, came in at the second step by donating computing power to help build more advanced versions of the AI model. As part of the agreement, the researchers renamed the project, Valenzuela said (Originally, the verbose academic name was “High-Resolution Image Synthesis with Latent Diffusion Models”).
“Who are you betting on to deliver the future of creative AI? The world’s best AI researchers, artists, and software developers …or hedge fund promoters?”
The open source nature of Stable Diffusion has led to confusion over the model’s ownership that on one occasion spilled into the public sphere. In October, Runway released a new update to Stable Diffusion, prompting Stability to issue a takedown request over IP rights. “They are bad faith actors who agreed to one thing, didn’t get the consent of other researchers who worked hard on the project and then turned around and did something else,” Stability AI’s CIO Daniel Jeffries wrote in a post on internet forum Reddit. Stability ultimately retracted its takedown request, with CEO Emad Mostaque telling Forbes at the time that he had wanted more guardrails against bad actors in place prior to releasing the model.
“Who are you betting on to deliver the future of creative AI? The world’s best AI researchers, artists, and software developers …or hedge fund promoters?” wrote Amplify Partners founder Sunil Dhaliwal, a Runway board member, on Twitter (Stability’s Mostaque was previously a hedge fund manager). For his part, Valenzuela clarified that Runway does not have an official partnership with Stability. Runway has since stepped away from core involvement with Stable Diffusion’s newer releases, including version 2.0 which was released in late November, according to Valenzuela. “I’m not really sure what Stable Diffusion 2.0 is,” he told Forbes Monday.
In any case, after the release of Stable Diffusion, both companies have become benefactors of the same hype in generative AI. Forbes reported that Stability raised at a $1 billion valuation, while others like Jasper ($1.5 billion) and Descript ($550 million, according to a report from The Information) have also raised large rounds in recent months. In November, before Forbes learned of the funding, Valenzuela alluded to the flock of familiar and new VC faces that have reached out to him.
“Everyone now is like, ‘I always thought it was going to work and I want to invest,’” he said. “But three years ago they had no conviction on this. It’s a great way for me to know who is being opportunistic versus deeply understanding our technology.”
Alex Konrad contributed reporting.