Downloads · 30 days
0
pravsels/pi05-so101-stacking-rings-isambard
pi05-so101-stacking-rings-isambard is a robotics model from pravsels. 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 lerobot. The card lists the license as apache-2.0.
Fine-tuned pi0.5 checkpoint for the SO101 stacking-rings task (6D joint-space, delta joints + absolute gripper). Trained on Isambard GH200 (4× GPU node); replication of the GCloud run.
Downloads · 30 days
0
Access
Public
Updated Jun 3, 2026
Repo size
24.9 GB
Likes
0
Public
Click a slice to open those files.
Other12.4 GB · 100%
From the Hugging Face model README
Fine-tuned pi0.5 checkpoint for the SO101 stacking-rings task (6D joint-space, delta joints + absolute gripper). Trained on Isambard GH200 (4× GPU node); replication of the GCloud run.
pi05_so101_stacking_rings fine-tune on Isambard.pi05_so101_stacking_ringspi05=True, action_horizon=30)[T,T,T,T,T,F] · Default prompt: stack the ringscd checkpoints/49999/params && find . -type f | sort | xargs sha256sum | sha256sum
| Step | Loss (approx) | SHA-256 |
|---|---|---|
| 49,999 | 0.0058 | d2b3cc68c91e6801fc81730f53be87e209cfe1e7b5932ca6bbdf6dd74c788f5d |
Signed MODEL_PASSPORT.json + SIGNOFF.json at repo root (checkpoint-passport v0.2).
from openpi.training.config import get_config
from openpi.serving.policy_server import PolicyServer
config = get_config("pi05_so101_stacking_rings")
server = PolicyServer(config, checkpoint_path="checkpoints/49999/params")