Instructions to use DecentVLA/pi05_cubestack_cent_3client with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LeRobot
How to use DecentVLA/pi05_cubestack_cent_3client with LeRobot:
- Notebooks
- Google Colab
- Kaggle
pi0.5 CubeStack โ Centralized (3-client union)
Centralized ceiling: full fine-tune on the POOLED union of all 6 tasks (2 per arm across harry/zhekai/kevin), 22500 steps. Shared pooled-6-repo normalizer. The upper bound for the 3-client local < FL < cent comparison.
lerobot-native export (portable tokenizer, real 6-D action/state; --policy.compile_model=false).
SO-101 CubeStack, cameras base_0_rgb/left_wrist_0_rgb. Trained on Isambard-AI (GH200)
with decent-vla. Part of the 3-client pi0.5
CubeStack federated study (color-pair non-IID: c0=harry Green/Orange, c1=zhekai Green/Blue,
c2=kevin Orange/Blue; each client blind to the 3rd color).
- Downloads last month
- 22
Model tree for DecentVLA/pi05_cubestack_cent_3client
Base model
lerobot/pi05_base