About

About GEN-0

GEN-0 is Generalist’s embodied foundation-model generation for multimodal training on raw physical interaction. Harmonic Reasoning interleaves asynchronous, continuous-time streams of sensing and acting tokens, allowing the model to coordinate perception and control across different robot embodiments.

Generalist tested GEN-0 on 6-DoF, 7-DoF and 16-plus-DoF semi-humanoid robots and scaled the family beyond 10 billion parameters. The company reports pretraining on more than 270,000 hours of real-world manipulation data spanning homes, warehouses and workplaces, with downstream tasks including assembly, packing, clothing handling and food preparation.

Timeline

Generalist gives first public live demo of GEN-0 at NVIDIA GTC
2026-03-24 - Demonstration - Generalist said its GEN-0 model ran in its first public live demo during NVIDIA GTC on a new mobile-manipulation system built from UR7e arms, a MiR base, and a Vention frame. The company said it had only a handful of days to prepare on the new hardware yet ran the demo throughout open exhibit hours without using data collected inside the GTC hall. Generalist framed the result as evidence that its foundation model could generalize quickly to new robots and new environments.
Source
Generalist introduces GEN-0 embodied foundation model
2025-11-04 - Product Launch - Generalist introduced GEN-0 as a new class of embodied foundation model trained directly on high-fidelity physical interaction data. The company said GEN-0 had scaled past 10B parameters, showed a phase transition around 7B in robotics pretraining, and worked across 6DoF, 7DoF, and 16+DoF semi-humanoid systems. Generalist also disclosed that its in-house pretraining corpus had reached more than 270,000 hours of real-world manipulation data and was growing by over 10,000 hours per week.
Source