Downloads · 30 days
6
7% of all-time downloads
so101-pen-in-box-v2-policy
c299m/so101-pen-in-box-v2-policy
so101-pen-in-box-v2-policy is a machine learning model from c299m. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Task: Pick up the pen and place it in the box Embodiment: GR1 (singlearm + gripper) Modalities: front camera, wrist camera, robot states/actions (5 DoF arm + gripper)
All-time downloads
86
Public
Parameters
2.7B
64.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors37.9 GB · 53%
Parameter types
How the weights are stored.
BF161.7B · 61%
Base models
Model card
From the Hugging Face model README
so101-pen-in-box-v2-policy
Task: Pick up the pen and place it in the box
Embodiment: GR1 (single_arm + gripper)
Modalities: front camera, wrist camera, robot states/actions (5 DoF arm + gripper)
📦 Checkpoints
| Step | Subfolder | Notes |
|---|---|---|
| 3000 | checkpoints/step-3000 | early |
| 5000 | checkpoints/step-5000 | mid |
| 7000 | checkpoints/step-7000 | mid |
| 10000 | checkpoints/step-10000 | latest (recommended) |
🗂️ Data
- Dataset:
c299m/so101-pen-in-box-v2 - Modality config:
so100_dualcam(front + wrist) - Annotation: single task description: “Pick up the pen and place it in the box.”
🚀 Quickstart
from transformers import AutoModel # replace with concrete class if needed
repo_id = "c299m/so101-pen-in-box-v2-policy"
sub = "checkpoints/step-10000" # or step-3000/5000/7000
model = AutoModel.from_pretrained(repo_id, subfolder=sub)
model.eval()
- Type
- gr00t_n1_5
- License
- apache-2.0
- Languages
- ja
- Created
- Aug 6, 2025
- Updated
- Aug 6, 2025