Generalist, a startup focused on robotics, has recently achieved a valuation of $3 billion following a new funding round led by 8VC, as confirmed by two sources familiar with the investment details.
The latest funding amounts to nearly $200 million, according to a regulatory disclosure. This new investment is part of an extended Series B round that initially raised $400 million, led by Radical Ventures, which the company initially announced in June at a valuation of $2 billion. With this latest addition, the total investment for this round has reached $600 million.
Neither Generalist nor 8VC has responded to inquiries for comments on the funding news.
Founded in 2024 by former Google DeepMind researchers Pete Florence and Andy Zeng, along with ex-Boston Dynamics engineer Andrew Barry, Generalist received early investment from notable firms including 8VC, Radical Ventures, Nvidia, Union Square Ventures, Bezos Expeditions, and prominent AI researcher Fei-Fei Li.
Until recently, Generalist maintained a low profile with minimal public exposure.
The startup is focused on creating an AI foundation model capable of operating a variety of robots. It recently introduced its Gen 1.5 model, which purportedly enables robots to learn new tasks from video clips ranging from 3 to 12 seconds in duration.
Generalist is collaborating with a select group of clients, utilizing their insights to customize the model for particular applications, according to one insider.
However, Generalist is not the only company aiming to develop advanced solutions for a diverse array of robotic applications. Competitors in the field include Physical Intelligence, which is valued at around $11 billion, and Skild AI, supported by SoftBank, with a valuation of about $14 billion, along with Genesis AI, which was reportedly negotiating for new funding at a $3 billion valuation as recently as last month.
The influx of capital signals a belief among certain investors that robotics could soon experience its own "ChatGPT moment," where robots will be able to perform a variety of general tasks without needing specific training for each one. Nevertheless, some venture capitalists caution that, unlike large language models trained on vast amounts of internet data, the creation of a truly generalized robotics model may still be years away.
