{
  "schema_version": "openagent.resource.v1",
  "id": "res_isaac_gr00t",
  "slug": "isaac-gr00t",
  "status": "published",
  "identity": {
    "name": "NVIDIA Isaac GR00T",
    "one_liner": "Open foundation model for generalist humanoid robots — VLA with real-time whole-body control.",
    "short_description": "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."
  },
  "classification": {
    "resource_type": "bot",
    "primary_category": "bots",
    "subcategories": [
      "robotics",
      "vla",
      "humanoid",
      "deep-learning",
      "gpu"
    ]
  },
  "positioning": {
    "why_it_matters": "GR00T matters because it is one of the first open foundation models capable of whole-body humanoid control — legs, arms, and hands coordinated from a single VLA policy. By open-sourcing model weights, fine-tuning code, and evaluation benchmarks under Apache 2.0, NVIDIA gives the robotics community a production-grade starting point for generalist robot intelligence that previously required millions of dollars and proprietary datasets.",
    "best_for": [
      "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"
    ],
    "not_for": [
      "Teams looking for a full robotics SDK (GR00T is a model, use Isaac Lab or Isaac Sim for the full stack)"
    ],
    "use_cases": [
      "robotics-agent"
    ],
    "target_audience": [
      "developer",
      "agent_builder"
    ],
    "maturity": "active"
  },
  "decision_signals": {
    "deployment_modes": [
      "cloud"
    ],
    "open_source": true,
    "local_first": false,
    "self_hostable": false,
    "has_api": false,
    "has_gui": false,
    "supports_mcp": false,
    "supports_docker": false
  },
  "facts": {
    "license": "Apache-2.0",
    "pricing_model": "open_source",
    "github_stars": 7236,
    "github_forks": 1232,
    "github_repo_full_name": "NVIDIA/Isaac-GR00T",
    "last_verified_at": "2026-06-04"
  },
  "capabilities": {
    "core_capabilities": [
      "robotics",
      "messaging"
    ],
    "interfaces": [
      "repo",
      "docs"
    ]
  },
  "links": {
    "primary_url": "https://github.com/NVIDIA/Isaac-GR00T",
    "items": [
      {
        "type": "github",
        "label": "GitHub",
        "url": "https://github.com/NVIDIA/Isaac-GR00T"
      },
      {
        "type": "homepage",
        "label": "Homepage",
        "url": "https://developer.nvidia.com/isaac/gr00t"
      }
    ]
  },
  "media": {
    "thumbnail_brief": {
      "resource_type": "bot",
      "visual_motif": "humanoid robot diagram with neural network overlay and joint trajectory",
      "background_style": "quiet editorial card with light surface and green accent",
      "title_overlay": "Isaac GR00T",
      "subtitle": "Open foundation model for humanoid robots",
      "avoid": [
        "dark sci-fi robot imagery",
        "NVIDIA green screen aesthetic"
      ]
    }
  },
  "tags": {
    "category": [
      "bot",
      "open-source"
    ],
    "capability": [
      "robotics",
      "messaging"
    ],
    "constraint": [
      "open-source"
    ],
    "scenario": [
      "robotics-agent"
    ]
  },
  "relationships": {},
  "machine_readable": {
    "canonical_url": "https://www.openagent.bot/bots/isaac-gr00t",
    "json_url": "https://www.openagent.bot/bots/isaac-gr00t.json",
    "markdown_url": "https://www.openagent.bot/bots/isaac-gr00t.md"
  },
  "seo": {
    "title": "NVIDIA Isaac GR00T: Open VLA foundation model for generalist humanoid robots",
    "description": "NVIDIA Isaac GR00T is an open-source vision-language-action (VLA) foundation model for humanoid robots. Apache 2.0, fine-tunable, real-time inference."
  },
  "editorial": {
    "trust_note": "Verified from source links and project metadata.",
    "core_strengths": [
      {
        "title": "Whole-body humanoid control from a single VLA policy",
        "description": "GR00T predicts latent action tokens that a learned whole-body controller (GEAR-SONIC) decodes into coordinated leg, arm, and hand commands.",
        "why_it_matters": "Prior VLA models focused on arms only. GR00T extends to full humanoid control, enabling locomotion + manipulation from one model."
      },
      {
        "title": "Cross-embodiment zero-shot and fine-tuning",
        "description": "Pre-trained on diverse robot data including bimanual arms, semi-humanoids, and humanoids. Fine-tune on new embodiments with as few as 5 episodes.",
        "why_it_matters": "You can evaluate zero-shot on supported robots or adapt to a custom robot with minimal data collection."
      },
      {
        "title": "Commercial-friendly open license",
        "description": "Fully Apache 2.0 licensed with model weights, fine-tuning code, and evaluation benchmarks all publicly available.",
        "why_it_matters": "Open VLA models are rare. Apache 2.0 means startups and researchers can build commercial products without legal uncertainty."
      },
      {
        "title": "LeRobot dataset format compatibility",
        "description": "GR00T uses the LeRobot v2 dataset format, making it easy to use existing LeRobot datasets and tools for data preparation.",
        "why_it_matters": "Direct compatibility with the Hugging Face robotics ecosystem lowers the barrier to getting started."
      }
    ],
    "use_case_notes": [
      {
        "title": "Fine-tuning on custom robot hardware",
        "description": "Collect 5-50 demonstration episodes from your robot, convert to LeRobot format, and fine-tune GR00T for your specific embodiment and task."
      },
      {
        "title": "Whole-body humanoid deployment",
        "description": "Use GR00T with the SONIC controller on Unitree G1 or similar humanoids for tasks that require coordinated locomotion and manipulation."
      },
      {
        "title": "Benchmarking VLA models on LIBERO and SimplerEnv",
        "description": "Evaluate GR00T against standard benchmarks with provided fine-tuned checkpoints for Franka Panda and WidowX arms."
      }
    ],
    "compare_notes": [
      {
        "title": "Choose GR00T for whole-body humanoid VLA",
        "summary": "Most open VLA models (Pi0, Xiaomi Robotics-0) focus on manipulation only. GR00T uniquely supports whole-body control including locomotion.",
        "against": "arm-only VLA models"
      }
    ],
    "getting_started": [
      {
        "label": "View the GitHub repository",
        "url": "https://github.com/NVIDIA/Isaac-GR00T",
        "type": "github"
      },
      {
        "label": "NVIDIA Isaac GR00T homepage",
        "url": "https://developer.nvidia.com/isaac/gr00t",
        "type": "homepage"
      }
    ],
    "command_line": [
      {
        "label": "Clone and run GR00T inference",
        "command": "git clone https://github.com/NVIDIA/Isaac-GR00T && cd Isaac-GR00T && pip install -e .",
        "description": "Clone the repository and install dependencies. Follow the examples for zero-shot inference or fine-tuning."
      }
    ],
    "seo_article": {
      "intro": "NVIDIA Isaac GR00T represents a significant step toward open, generalist robot intelligence — a VLA foundation model that controls whole humanoid bodies rather than just arms.",
      "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_it_works": "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.",
      "use_cases": [
        {
          "title": "Humanoid manipulation and locomotion",
          "description": "Deploy GR00T on a Unitree G1 humanoid for tasks that require walking to a location, picking up an object, and placing it — all from a single language instruction."
        },
        {
          "title": "Fine-tuning for specialized industrial tasks",
          "description": "Collect demonstration data from a custom robot arm performing a specific assembly task, fine-tune GR00T, and deploy with TensorRT for real-time inference."
        },
        {
          "title": "Cross-embodiment robotics research",
          "description": "Use GR00T's pre-trained representations as a starting point for studying how VLA models transfer across different robot morphologies."
        }
      ],
      "alternatives": [
        {
          "title": "Use Pi0 or Pi0.5 for arm-only manipulation",
          "summary": "Physical Intelligence's Pi0 series excels at arm manipulation but does not support whole-body or locomotion control. GR00T is the choice when legs are involved.",
          "against": "GR00T"
        }
      ],
      "getting_started": [
        {
          "label": "Clone the repository",
          "url": "https://github.com/NVIDIA/Isaac-GR00T",
          "type": "github"
        },
        {
          "label": "NVIDIA developer portal",
          "url": "https://developer.nvidia.com/isaac/gr00t",
          "type": "homepage"
        }
      ],
      "faq": [
        {
          "question": "What hardware do I need to run GR00T?",
          "answer": "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."
        },
        {
          "question": "Can I use GR00T commercially?",
          "answer": "Yes, GR00T is fully licensed under Apache 2.0, including model weights, fine-tuning code, and evaluation tools."
        },
        {
          "question": "Does GR00T work with my robot?",
          "answer": "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."
        }
      ]
    }
  },
  "timestamps": {
    "created_at": "2026-06-04T00:00:00.000Z",
    "updated_at": "2026-06-04T00:00:00.000Z",
    "published_at": "2026-06-04T00:00:00.000Z"
  }
}