Upload folder using huggingface_hub
Browse files- .metadata.yaml +22 -0
- README.md +33 -0
.metadata.yaml
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# ROOT .metadata.yaml — "coming soon" placeholder.
|
| 2 |
+
# No `format` block and no artifact subfolders: nothing is published yet.
|
| 3 |
+
# The AI-Dashboard treats a repo with zero artifact folders as "coming soon"
|
| 4 |
+
# automatically — once real precision folders + benchmarks are added here,
|
| 5 |
+
# this becomes a normal model card with no other change needed.
|
| 6 |
+
|
| 7 |
+
model:
|
| 8 |
+
name: gr00t
|
| 9 |
+
display_name: GR00T
|
| 10 |
+
upstream: nvidia/GR00T-N1.5-3B # TODO: confirm exact upstream repo id/version before publishing
|
| 11 |
+
|
| 12 |
+
architecture:
|
| 13 |
+
family: gr00t
|
| 14 |
+
modality:
|
| 15 |
+
# GR00T is a humanoid-robot vision-language-action model — not a
|
| 16 |
+
# chat-style VLM. Renders on the dashboard with the VLA badge and
|
| 17 |
+
# illustration (assets/fallback-vla.svg), distinct from VLM.
|
| 18 |
+
- vision-language-action
|
| 19 |
+
parameters: 3.0B
|
| 20 |
+
|
| 21 |
+
tasks:
|
| 22 |
+
- robotics
|
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
base_model:
|
| 4 |
+
- nvidia/GR00T-N1.5-3B
|
| 5 |
+
pipeline_tag: robotics
|
| 6 |
+
tags:
|
| 7 |
+
- renesas
|
| 8 |
+
- x5h
|
| 9 |
+
- coming-soon
|
| 10 |
+
- vla
|
| 11 |
+
- robotics
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# GR00T – Renesas X5H
|
| 15 |
+
|
| 16 |
+
<p align="center">
|
| 17 |
+
<img src="https://raw.githubusercontent.com/NVIDIA/Isaac-GR00T/main/media/header_compress.png" alt="NVIDIA Isaac GR00T" width="720">
|
| 18 |
+
</p>
|
| 19 |
+
<p align="center">
|
| 20 |
+
<img src="https://media.githubusercontent.com/media/NVIDIA/Isaac-GR00T/main/media/unitree_g1.gif" alt="Isaac GR00T controlling a Unitree G1 humanoid robot" width="480">
|
| 21 |
+
</p>
|
| 22 |
+
<p align="center"><sub>Source: NVIDIA's official <a href="https://github.com/NVIDIA/Isaac-GR00T">Isaac-GR00T</a> repository</sub></p>
|
| 23 |
+
|
| 24 |
+
🚧 **Coming soon.** This model is being optimized and validated for Renesas hardware — no fixed release date yet.
|
| 25 |
+
|
| 26 |
+
## Introduction
|
| 27 |
+
|
| 28 |
+
**GR00T** is NVIDIA's generalist foundation model for humanoid robots — a vision-language-action (VLA) model that maps camera input and language instructions to robot actions. Renesas is preparing an optimized deployment of this model for the R-Car Gen5 platform.
|
| 29 |
+
|
| 30 |
+
- **Model Architecture:** Vision-language-action (VLA) model for robot control.
|
| 31 |
+
- **Source Model:** [nvidia/GR00T-N1.5-3B](https://huggingface.co/nvidia) *(exact upstream repo/version to be confirmed)*
|
| 32 |
+
|
| 33 |
+
*This page is a placeholder. Content — including the license and source-model link — is provisional and subject to change before the model is fully published.*
|