Generalist, a robotics startup founded in 2024, has raised about $200 million in a funding extension that values it at $3 billion, according to a regulatory filing reported by TechCrunch on 25 August. The new money, led by 8VC, builds on a Series B that opened in June at $400 million and a $2 billion valuation, both led by Radical Ventures, taking the round to roughly $600 million in total.
The company builds one artificial intelligence model meant to control many different robots rather than a single machine of its own. Co-founders Pete Florence and Andy Zeng came from Google DeepMind, and Andrew Barry from Boston Dynamics. Its latest release, Gen 1.5, is pitched as letting a robot learn a new task from video clips as short as three to twelve seconds. Backers include Nvidia, Union Square Ventures, Bezos Expeditions and the researcher Fei-Fei Li.
The valuation has tripled from the June mark in about ten weeks, with no product milestone tied to the raise. Generalist says it is working with a small number of customers and using their feedback to adapt the model to particular jobs. It has not named those customers or said what, if anything, they are paying for the work.
The robot foundation model is now a crowded idea. Google’s Gemini Robotics, Physical Intelligence and Skild are chasing the same claim, that a single model can generalise across different bodies and tasks. None has shown it holding up outside arranged demonstrations at a price a buyer will accept. Generalist’s $3 billion rests on that bet coming good rather than on evidence that it already has.
