Robotics startup Generalist reaches $3B valuation after new funding
Robotics startup Generalist has reached a $3 billion valuation following an additional funding round led by 8VC. The company is building a universal AI model designed to operate a wide range of robots.
Robotics startup Generalist has reached a $3 billion valuation after raising fresh capital in a round led by venture firm 8VC, according to two people familiar with the deal. A regulatory filing shows the new funding totals nearly $200 million.
The capital is an extension of a $400 million Series B round led by Radical Ventures, which the company announced in June at a $2 billion valuation. With the latest infusion, total funding raised in this round has climbed to $600 million. Neither Generalist nor 8VC responded to requests for comment.
Company background
Generalist was founded in 2024 by former Google DeepMind researchers Pete Florence and Andy Zeng, together with former Boston Dynamics engineer Andrew Barry. Early backers included 8VC, Radical Ventures, Nvidia, Union Square Ventures, Bezos Expeditions, and AI researcher Fei-Fei Li. Until recently, the company operated with little public visibility.
The startup is developing an AI foundation model capable of working across different types of robots. Its recently released Gen 1.5 model reportedly allows robots to learn new tasks from video demonstrations as short as 3 to 12 seconds. According to one source, Generalist is currently working with a small number of customers, using their feedback to fine-tune the model for specific applications.
A crowded field
Generalist is not alone in the race to build a general-purpose 'brain' for robots. Competitors include Physical Intelligence, reportedly valued at $11 billion, and SoftBank-backed Skild AI, valued at $14 billion. Another rival, Genesis AI, was reportedly in talks last month to raise funding at a $3 billion valuation.
The wave of investment reflects a belief among some investors that robotics could be approaching its own 'ChatGPT moment' — a point at which robots can perform general tasks without being specifically trained for each one. Still, some venture capitalists caution that a truly general robotics model may still be years away, since robots, unlike language models, cannot be trained on the entirety of internet data.
