- Robotics researchers fine-tuning VLA models for custom hardware
- Teams deploying humanoid robots with whole-body coordination
- Developers needing a commercially licensable open-source robot foundation model
Apache-2.0 · Bots
NVIDIA Isaac GR00T
Open foundation model for generalist humanoid robots — VLA with real-time whole-body control.
7.2K stars 1.2K forks Apache-2.0 license 2026-06-04 verified
$
git clone https://github.com/NVIDIA/Isaac-GR00T && cd Isaac-GR00T && pip install -e . Open source
Overview
What is NVIDIA Isaac GR00T?
NVIDIA Isaac GR00T is an open vision-language-action (VLA) foundation model family for generalized humanoid and manipulation robot skills. It takes multimodal input — language and images — and outputs joint-level action sequences for diverse robot embodiments. GR00T N1.7 supports zero-shot inference, fine-tuning on custom robot data, and real-time deployment with TensorRT acceleration. Built on the LeRobot dataset format and fully commercially licensable under Apache 2.0.
Whole-body humanoid control from a single VLA policy
GR00T predicts latent action tokens that a learned whole-body controller (GEAR-SONIC) decodes into coordinated leg, arm, and hand commands.
Prior VLA models focused on arms only. GR00T extends to full humanoid control, enabling locomotion + manipulation from one model.Cross-embodiment zero-shot and fine-tuning
Pre-trained on diverse robot data including bimanual arms, semi-humanoids, and humanoids. Fine-tune on new embodiments with as few as 5 episodes.
You can evaluate zero-shot on supported robots or adapt to a custom robot with minimal data collection.Commercial-friendly open license
Fully Apache 2.0 licensed with model weights, fine-tuning code, and evaluation benchmarks all publicly available.
Open VLA models are rare. Apache 2.0 means startups and researchers can build commercial products without legal uncertainty.LeRobot dataset format compatibility
GR00T uses the LeRobot v2 dataset format, making it easy to use existing LeRobot datasets and tools for data preparation.
Direct compatibility with the Hugging Face robotics ecosystem lowers the barrier to getting started. Install
One command to start
$
git clone https://github.com/NVIDIA/Isaac-GR00T && cd Isaac-GR00T && pip install -e . Use cases
What teams use it for
Ecosystem
Tags & capabilities
Comparison
How it stacks up
Choose GR00T for whole-body humanoid VLA
vs arm-only VLA modelsMost open VLA models (Pi0, Xiaomi Robotics-0) focus on manipulation only. GR00T uniquely supports whole-body control including locomotion.
FAQ
Questions
What hardware do I need to run GR00T?
GR00T requires a GPU with sufficient VRAM for the 3B parameter model. A single A100 or RTX 6000 Ada is recommended for training. For inference, TensorRT-optimized deployment runs on consumer GPUs.
Can I use GR00T commercially?
Yes, GR00T is fully licensed under Apache 2.0, including model weights, fine-tuning code, and evaluation tools.
Does GR00T work with my robot?
GR00T supports multiple embodiments through its embodiment tag system. Supported tags include LIBERO PANDA, DROID, SO100, SimplerEnv, and UNITREE G1 SONIC. New embodiments can be added via fine-tuning.
Decision brief
JSON
Should you use NVIDIA Isaac GR00T?
- Teams looking for a full robotics SDK (GR00T is a model, use Isaac Lab or Isaac Sim for the full stack)
- Verified 2026-06-04
- License: Apache-2.0
- Repo: NVIDIA/Isaac-GR00T
- Open-source signal
Deployment
cloud
Permission surface
messages, hardware
Decision signals
No extra signals recorded
Agent packet
Structured decision data for NVIDIA Isaac GR00T
This packet is the compact machine-readable view agents should use before following source links or taking action.
Capabilities
robotics, messaging
Constraints
open source
Deployment
cloud
Permission surface
messages, hardware
Recommended workflows
Robotics or embodied agent workflow
Overview
What NVIDIA Isaac GR00T does
What it is
GR00T N1.7 is a 3-billion parameter vision-language-action model that takes language instructions and camera images as input and outputs latent action tokens. A learned whole-body controller (SONIC) decodes these tokens into coordinated joint-level commands for legs, arms, and hands. The model is pre-trained on a diverse mixture of robot data including bimanual manipulation, semi-humanoid, and humanoid datasets, plus 20K hours of human video data.
Why it matters
Generalist robot models have been dominated by closed-source efforts. GR00T changes that by open-sourcing a production-quality VLA under Apache 2.0. For the first time, a researcher with a humanoid robot can download a foundation model, fine-tune it on their specific tasks, and deploy it commercially — all without NVIDIA licensing constraints. This accelerates the whole field of physical AI.
How to evaluate it
Data collection uses the LeRobot v2 format with synchronized video and action sequences. The model is fine-tuned via a launch_finetune.py script that handles modality configuration. For deployment, Gr00tPolicy connects to the robot controller, optionally accelerated with TensorRT. The model supports both zero-shot inference on pre-trained embodiments and fine-tuned deployment for custom robots.
Facts
Known metadata and operating surface
These fields are separated from editorial interpretation so agents can reason over facts and missing checks.
Resource type bot
Category Bots
Maturity active
Difficulty Unknown
License Apache-2.0
Pricing open source
Verified 2026-06-04
Source confidence high
Risk level elevated
Fit matrix
Where NVIDIA Isaac GR00T fits in an agent stack
strong
Robotics or embodied agent workflow
NVIDIA Isaac GR00T has multiple signals for robotics or embodied agent workflow, including matching tags, capabilities, category, or positioning.
- Separate simulator claims from hardware claims and verify safety boundaries before real-world operation.
- Confirm official docs, current maintenance, license, and runtime constraints before production use.
partial
Browser automation
NVIDIA Isaac GR00T has at least one signal for browser automation, but should be checked against a real task before adoption.
- Run one non-sensitive website task and inspect clicks, waits, retries, and changed URLs.
- Confirm official docs, current maintenance, license, and runtime constraints before production use.
partial
Coding agent workflow
NVIDIA Isaac GR00T has at least one signal for coding agent workflow, but should be checked against a real task before adoption.
- Run a small repository change and inspect the diff, tests, and rollback path.
- Confirm official docs, current maintenance, license, and runtime constraints before production use.
partial
Connector or protocol layer
NVIDIA Isaac GR00T has at least one signal for connector or protocol layer, but should be checked against a real task before adoption.
- Connect one low-risk service, then inspect schemas, auth scope, errors, and logs.
- Confirm official docs, current maintenance, license, and runtime constraints before production use.
partial
Reusable skill workflow
NVIDIA Isaac GR00T has at least one signal for reusable skill workflow, but should be checked against a real task before adoption.
- Run one skill end to end and check whether it produces evidence or structured output.
- Confirm official docs, current maintenance, license, and runtime constraints before production use.
weak
Evaluation and observability
NVIDIA Isaac GR00T is not primarily positioned for evaluation and observability in the current metadata.
- Add one repeatable test case and confirm results can run again in review or CI.
- Confirm official docs, current maintenance, license, and runtime constraints before production use.
Inputs and outputs
What an agent should inspect
Likely inputs
- Repositories, files, issues, terminal output, and test results
- Prompts, messages, documents, images, or model inputs
- Official setup instructions and a small real workflow
Likely outputs
- Diffs, commits, explanations, test results, or review notes
- A decision on whether this resource fits the target workflow
Evidence
Sources, claims, and missing checks
Claims are marked separately from source links so future crawlers and reviewers can update them without rewriting the page.
GitHub github
Repository source for code, license, issues, releases, and implementation details.
Homepage homepageOfficial or project-controlled source for this resource profile.
NVIDIA Isaac GR00T is listed as open source.
License metadata: Apache-2.0NVIDIA Isaac GR00T has a recorded GitHub repository: NVIDIA/Isaac-GR00T.
Resource facts and GitHub source link.NVIDIA Isaac GR00T supports these recorded deployment modes: cloud.
OpenAgent decision signal metadata.NVIDIA Isaac GR00T is tagged with robotics, messaging capabilities.
OpenAgent capability taxonomy.- Dedicated docs link is missing.
- Repository freshness has not been recorded.
Next action
How to start evaluating NVIDIA Isaac GR00T
Inspect repository
Check license, recent activity, issues, examples, and security-sensitive code paths.
Open sourceOpen Homepage
Start from the official source before adopting third-party instructions.
Open sourceClone and run GR00T inference
Clone the repository and install dependencies. Follow the examples for zero-shot inference or fine-tuning.
git clone https://github.com/NVIDIA/Isaac-GR00T && cd Isaac-GR00T && pip install -e . Compare
Alternatives and nearby resources
Use related resources to compare category fit, license, deployment model, and first-workflow behavior.
FAQ
Common questions about NVIDIA Isaac GR00T
What hardware do I need to run GR00T?
GR00T requires a GPU with sufficient VRAM for the 3B parameter model. A single A100 or RTX 6000 Ada is recommended for training. For inference, TensorRT-optimized deployment runs on consumer GPUs.
Can I use GR00T commercially?
Yes, GR00T is fully licensed under Apache 2.0, including model weights, fine-tuning code, and evaluation tools.
Does GR00T work with my robot?
GR00T supports multiple embodiments through its embodiment tag system. Supported tags include LIBERO PANDA, DROID, SO100, SimplerEnv, and UNITREE G1 SONIC. New embodiments can be added via fine-tuning.