Downloads · 30 days
0
droid-cube-stacking-finetune-e20000
jaehyeokdoo2/droid-cube-stacking-finetune-e20000
droid-cube-stacking-finetune-e20000 is a robotics model from jaehyeokdoo2. Use it for the robotics task on the model card, and read the license before you ship it in a product. It is set up for openpi.
Access
Public
Updated Mar 10, 2026
Parameters
3.6B
20.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt13.5 GB · 64%
Parameter types
How the weights are stored.
BF163.5B · 97%
Model card
From the Hugging Face model README
OpenPI Fine-tuned Model
This model was fine-tuned using OpenPI.
Training Details
- Global Step: 20000
- Experiment Name: droid-finetune-cubestacking
- Learning Rate: N/A
- Batch Size: 32
Usage
from openpi.models_pytorch.pi0_pytorch import PI0Pytorch
from safetensors.torch import load_file
from huggingface_hub import hf_hub_download
# Download model weights
model_path = hf_hub_download(repo_id="your-repo-id", filename="model.safetensors")
state_dict = load_file(model_path)
# Load model (adjust config as needed)
model = PI0Pytorch(...)
model.load_state_dict(state_dict)
- Task
- Robotics
- Library
- openpi
- Created
- Mar 10, 2026
- Updated
- Mar 10, 2026