Generalist, a startup building AI foundation models that can work across different robots, has reached a $3 billion valuation after quietly raising nearly $200 million in an extension of its Series B, according to two people with knowledge of the funding.
The new capital, first reported by Axios on August 24, brings the round’s total to $600 million. 8VC led the extension, with existing investors participating. The company had announced a $400 million Series B in June at a $2 billion valuation, led by Radical Ventures.
That is a steep trajectory for a company that was founded in 2024 and operated largely in stealth until recently.
What Generalist Is Building
Generalist is not building robot hardware. It builds the AI models that serve as the “brain” for robots — the equivalent of what large language models are for chatbots, but for physical manipulation and navigation.
The company says its Gen 1.5 model enables robots to learn new tasks from video demonstrations as short as 3 to 12 seconds. That matters because the bottleneck in robotics has never been building the hardware — it has been teaching the hardware to do useful things without months of programming for each specific task. If a robot can watch a short clip and figure out how to pick up a new object or navigate a new space, the economics of deployment shift dramatically.
Founded by former Google DeepMind researchers Pete Florence and Andy Zeng, along with former Boston Dynamics engineer Andrew Barry, Generalist sits at the intersection of two trends: the maturation of vision-language-action models that combine perception, reasoning, and physical control, and the surge in humanoid robot shipments that has seen Chinese manufacturers dominate global output.
The company’s backers include Nvidia, Union Square Ventures, Bezos Expeditions, AI researcher Fei-Fei Li, and Spark Capital. Generalist and 8VC did not respond to requests for comment from either Axios or TechCrunch.
A Crowded and Expensive Field
Generalist is far from alone in the race to build a universal robot brain, and the valuations are getting eye-watering.
Physical Intelligence is reportedly valued at $11 billion. SoftBank-backed Skild AI sits at $14 billion. Genesis AI was reportedly in talks last month to raise at a $3 billion valuation. These are companies building software, not selling robots — yet the money flowing into them suggests investors believe the payoff from a general-purpose robotics model could rival or exceed what we have seen with language models.
The analogy investors are making is explicit: a “ChatGPT moment” for robotics, where robots can perform general tasks without being explicitly trained for each one. That phrase has been circulating since Unitree’s CEO used it earlier this year, and it has become the shorthand for what every robotics AI company is promising.
The sceptical counter, raised by some VCs in TechCrunch’s reporting, is that robots cannot be trained on the entire internet the way language models can. Physical data is scarce, expensive to collect, and varies enormously across robot morphologies. A truly general robotics model may still be years away — which raises the question of whether these valuations are pricing in a breakthrough that has not happened yet.
Why This Matters for the Robotics Supply Chain
The humanoid robot industry has a split structure: companies like Unitree, Agibot, and Agility Robotics build the bodies, while companies like Generalist, Physical Intelligence, and Skild AI build the brains. The hardware side is concentrated in China, which shipped over 22,000 humanoid robots in the first half of 2026. The software side is concentrated in the US, where the funding is flowing.
Whether these two sides converge — whether a single company eventually builds both the body and the brain — is one of the strategic questions that will shape the industry over the next two years. For now, the money is betting that the brain is worth more than the body.
That bet is getting more expensive by the month.
❓ FAQ
What is Generalist? A robotics AI startup founded in 2024 by former Google DeepMind researchers and a former Boston Dynamics engineer. It builds AI foundation models designed to work across different types of robots, rather than building robot hardware itself.
How much has Generalist raised? The company has raised approximately $600 million in total, including a $400 million Series B in June 2026 and a roughly $200 million extension in August 2026. Its latest valuation is $3 billion, up from $2 billion in June.
Who are Generalist’s competitors? Physical Intelligence (reportedly $11 billion valuation), Skild AI ($14 billion, SoftBank-backed), and Genesis AI (in talks for a $3 billion round). All are building general-purpose AI models for robotics.
Why are investors betting on robot brains rather than robot bodies? The logic is that a general-purpose robotics model — one that works across different hardware — could become as foundational as a large language model, capturing value across the entire robotics industry rather than from a single robot platform.
