Generalist, a Bay Area and Boston-based developer of embodied foundation models for general-purpose robots, has raised $400 million in funding led by Radical Ventures. The company builds GEN-0 and GEN-1 models that enable dexterous physical tasks across robot types. The capital will accelerate scaling of physical AI.
Physical AI Funding Frenzy Accelerates
The timing comes amid surging investment in embodied AI. Physical Intelligence is in talks to raise $1 billion at an $11 billion-plus valuation. Figure AI has raised over $1.9 billion. Generalist's approach, training models from scratch on real-world manipulation data rather than extending video-language-action models, targets the gap in reliable physical commonsense and closed-loop reactivity.
Dexterity Limits Robot Deployment
Current robots struggle with dexterous manipulation in unstructured environments, limiting adoption in factories, warehouses, and homes. Labor shortages drive demand, yet most systems fail on variable tasks. Generalist reports GEN-1 achieving 99% success rates on diverse tasks versus prior 64%, using one hour of robot data per result after pretraining on 500,000 hours of real-world data.
Scaling Laws Applied to Physical Tasks
Generalist built GEN-0 and GEN-1 as foundation models that generalize across hardware embodiments. The models input raw video pixels and output actions, achieving emergent improvisational intelligence. This differs from competitors like Covariant, which specializes in warehouse picking with $222 million raised, or 1X Technologies, focused on home humanoids with $275 million-plus.
"General intelligence will be born from the physical world."
Radical Ventures Leads With AI Thesis
Radical Ventures led the round, with participation from Nvidia, Bezos Expeditions, 8VC, Union Square Ventures, and Hanabi Capital. The investor mix signals conviction in physical AI as the next scaling frontier, with Nvidia providing compute validation and Bezos Expeditions adding frontier-tech alignment. Total funding now exceeds $500 million at a $2 billion post-money valuation.
Market Projects 16x Growth by 2035
The physical AI market is projected to grow from $5.23 billion in 2025 to $87.43 billion by 2035 at 32.53% CAGR. Broader AI agents market stands at $11.55 billion in 2026 with 43-46% CAGR. Capital inflows reflect proof that scaling laws extend to robotics, with multiple players raising hundreds of millions in 2026.
Data Flywheel Begins to Turn
With deployments moving to early access partners and live demos at NVIDIA GTC, Generalist is generating proprietary interaction data that improves model performance. The company positions this flywheel as the differentiator enabling mastery on simple physical tasks.