molmoact2_fold_tea_towel

Fine-tuned MolmoAct2 (action-expert-only) on SO101 data.

Base model	allenai/MolmoAct2
Dataset	villekuosmanen/armnetbench_fold_tea_towel
Task	fold_tea_towel
Training	Vast.ai H100 SXM, batch 64, bf16, gradient checkpointing

Checkpoints

Step	Path
006000	checkpoints/006000/pretrained_model/
004000	checkpoints/004000/pretrained_model/
002000	checkpoints/002000/pretrained_model/

Usage

from lerobot.policies.molmoact2.modeling_molmoact2 import MolmoAct2Policy
policy = MolmoAct2Policy.from_pretrained("pravsels/molmoact2_fold_tea_towel", revision="main")