Automate tabletop manipulation tasks for SO-100 and SO-101 follower arms.
npx skills add https://github.com/bensonlee5/openral --skill molmoact2-so101-nf4
Please help me install this Agent Skill. Skill: molmoact2-so101-nf4 Source: https://github.com/bensonlee5/openral/tree/main/rskills/molmoact2-so101-nf4 Command: npx skills add https://github.com/bensonlee5/openral --skill molmoact2-so101-nf4
Automates tabletop manipulation tasks like picking, placing, and grasping with the SO-100/101 follower arms.
Use the molmoact2-so101-nf4 skill with the SO-101 box scene to execute actions on the robot.
High-intent search queries and answers about installing and using this skill.
Automate tabletop manipulation tasks for SO-100 and SO-101 follower arms by leveraging MolmoAct2's vision-language-action policy to generate continuous joint-position actions for arm movements.
MolmoAct2's vision-language-action policy is a continuous joint-position action mechanism that drives robotic arm movements, specifically designed to automate tabletop manipulation tasks like picking and placing objects.
Yes, you need OpenRAL integration for model loading and execution when using the MolmoAct2 action policy to automate manipulative tasks on SO-100 and SO-101 follower arms.
Yes, you can use this approach to execute tabletop manipulation tasks such as picking, placing, and grasping objects directly with the SO-100 and SO-101 follower arms.
Yes, the Skill supports both SO-100 and SO-101 follower arms, providing a continuous joint-position action policy to automate manipulative tasks across both robotic platforms.
The NF4 implementation provides efficient manipulation actions for SO-100 and SO-101 arms, though researchers must ensure proper OpenRAL integration for model loading and execution within their tabletop environments.
bensonlee5Automate pick-and-place 'pass the pen' tasks on a real SO-101 arm.
bensonlee5Automate pen pick-and-place tasks on a real SO-101 follower arm using a VLA policy.
bensonlee5Automate pick and place actions on tabletop objects using a VLA policy.
bensonlee5Automate Franka Panda pick, place, open, and close actions using a vision-language-action policy.
