Nvidia's Robot Models Rely on Pre-training
chris_j_paxton · x · 2026-07-14
This post conveys a perspective: Nvidia's open-source models for robotics aren't trained entirely from scratch; their underlying pre-training still relies on models/architectures from Alibaba.
The author uses this to highlight the significant mutual dependency already existing in the robotics model ecosystem between China and the US, while also calling for more "from-scratch pre-trained" foundational models for robotics.
More from Embodied
- NVIDIA brings its Cosmos 3 Edge world model to Jetson for on-device robot control — liu_mingyu · 2026-07-21
- A set of agent skills for CAD, robotics, and hardware design — earthtojake · 2026-07-21
- DIY wooden box packs 6 Intel Arc Pro B70 cards with a FreeCAD model — nick_ziv · 2026-07-21
- Snake-like robot moves on fully passive wheels and winding motion — ___Mufasaa · 2026-07-21
- Creator buys a Reachy robot and asks what to build first — dee_hw · 2026-07-21
- MW team shows Gen1 of MW-bot, a semi-humanoid home robot built for pantry storage and ceiling rails — CyberRobooo · 2026-07-21