What exactly happened

In early July 2026, Nvidia published details on its official blog about its collaboration with Hugging Face, which brings key technologies for physical AI into the open robotics library LeRobot. LeRobot functions as a "Hugging Face for robots" — a repository of models, datasets, and tools that anyone can use, modify, and share.

Specifically, the integration rests on three main pillars:

• Isaac GR00T 1.7 — an open VLA model (Vision-Language-Action) that understands images and speech and can generate actions for humanoid robots. It is the first model of its kind with a commercially usable license.
• Isaac Teleop — a tool for collecting demonstration data from human operators. It enables standardized recording of movements and the creation of training datasets.
• Cosmos 3 (planned) — an advanced "world model" capable of generating realistic training data and simulating scenarios when real-world data is expensive or unavailable.

Why it matters: an open-source model swimming against the current

While most major AI labs lock their best models behind APIs and hefty licensing fees, Nvidia is betting on the opposite strategy. Isaac GR00T 1.7 is fully open-source and available under a license that permits commercial use. This is a fundamental difference from competing robotics platforms by OpenAI or Google DeepMind, which remain proprietary.

"Open source is the way to turn advanced research into something people can study, adapt, and build upon," said Thomas Wolf, co-founder and Chief Science Officer of Hugging Face. And it's not just a platitude — Nvidia is already providing the largest open dataset for physical AI with more than 350,000 trajectories and 57 million grasps, which has been downloaded over 15 million times.

What a VLA model means and how it works

VLA (Vision-Language-Action) is a type of AI model that combines three capabilities: seeing the environment through a camera (vision), understanding natural language instructions (language), and generating physical robot movements based on that input (action). In practice, this means you can simply tell a robot "hand me that blue screwdriver" — and GR00T 1.7 will evaluate the scene, identify the object, and execute the necessary grasp.

Dataset, simulation, and deployment: the entire development pipeline in one package

The integration goes far beyond the model itself. Nvidia is also adding to the LeRobot ecosystem:

• Isaac Sim and Isaac Lab — simulation environments for testing robotic policies before deploying them on physical hardware. A developer can fine-tune everything virtually first.
• Isaac Lab-Arena connected to the LeRobot Environment Hub — enabling prototyping of complex simulation environments and immediately using them for training models like GR00T, Pi, or SmolVLA.
• Jetson Thor — a computing module integrated with the Reachy 2 humanoid robot, enabling deployment of VLA models directly on open-source hardware. No cloud, no API fees — everything runs locally.

For Czech developers and small businesses, this is a crucial message: the barrier to entry in robotics is dropping dramatically. While a year ago developing a humanoid robot would have required a team of dozens of specialists and millions in investment, today even a talented student at CTU or BUT with access to open-source hardware can "program a robot."

Czech context: robotics as a national sport

The Czech Republic has a historic connection to robotics — from the word "robot," coined by Karel Čapek, to a strong industrial base in automation. Companies like ABB (robotics development in Brno), Škoda Auto (automated production lines), and a growing startup scene in Prague and Brno incubators can significantly benefit from Nvidia's open approach.

While large language models like ChatGPT or Claude support Czech well, the situation with robotic VLA models is more complex — GR00T 1.7 is primarily trained on English. However, thanks to the open LeRobot ecosystem, it can be fine-tuned for Czech instructions using custom datasets.

Nvidia vs. the rest of the world: who will win the race for a robotics platform

Nvidia's strategy is clear: become the layer on which all physical AI development runs — much like CUDA became the standard for training neural networks. But the competition is not standing still:

• Google DeepMind is developing RT (Robotics Transformer) models, but they remain behind corporate walls.
• OpenAI has introduced its own robotics division, but it does not publicly share specific models yet.
• Agibot (a Chinese startup backed by Alibaba) is forging ahead with commercial humanoid robots, but with closed software.

Nvidia's openness may be what tips the scales in its favor — especially in Europe, where regulations like the EU AI Act require transparency and explainability of systems. An open-source model whose code you can inspect has a significantly easier position before regulators than a black box from the competition.

What comes next: Cosmos 3 and the end of training data scarcity

The most intriguing part of the announcement is yet to come. Nvidia Cosmos 3 — a "world foundation model" for physical AI — is set to arrive in the LeRobot library in the coming weeks. It is a model capable of generating realistic training data for robotic scenarios. In practice, this means that instead of manually recording thousands of movement demonstrations, a developer can have Cosmos 3 generate synthetic data that closely approximates reality.

For smaller robotics labs — including academic ones like CIIRC CTU or CEITEC BUT — this could mean a radical acceleration of research. No more weeks of manual data collection, but hours of synthetic generation.

Summary: robotics for the masses

The partnership between Nvidia and Hugging Face within the LeRobot library represents a significant milestone for open-source robotics. For the first time in history, a commercially usable foundation model for humanoid robots is available that anyone can use, modify, and deploy for free. Add to that data collection tools, simulation in Isaac Sim, and the planned integration of Cosmos 3 — and you have an ecosystem that could make robotics as accessible a field as machine learning became after the arrival of libraries like PyTorch.

For the Czech Republic — a country with a rich industrial tradition and a growing AI scene — this is an opportunity that would be a shame to waste.