MolmoAct2: Action Reasoning Models for Real-world Deployment

MolmoAct2 is Ai2's open family of action reasoning models for robot control and real-world deployment. It builds on the Molmo2-ER embodied-reasoning vision-language backbone, adds robot state and action modeling, and connects the VLM to a flow-matching continuous action expert for closed-loop manipulation. The release includes base checkpoints for continued training, fine-tuned robot policies for evaluation and deployment, and the datasets used to build MolmoAct2 and Molmo2-ER.

We provide base checkpoints at every training stage for continued MolmoAct2 training and robot fine-tuning. These are foundation checkpoints rather than one-size-fits-all deployment policies.

We also provide fine-tuned checkpoints for common robot platforms and benchmarks. These models are intended to run directly in their target setting, or to serve as a stronger starting point for closely related robots. As with any robot policy, performance depends on hardware, cameras, calibration, action conventions, and language/task distribution.

Note that all of the robotics datasets for pre-training and post-training are in LeRobot v3.0 format, paired with extra language annotations.

The MolmoAct2 LeRobot fork is included as a Git submodule at lerobot/. After cloning this repository, initialize the submodule from the repo root:

git submodule update --init --recursive
cd lerobot

For LIBERO replication and other evaluation instructions, follow the local LeRobot README at lerobot/README.md after the submodule is initialized.

MolmoAct2 supports out-of-the-box deployment on three robot embodiments:

• SO-100
• Bimanual YAMs
• Franka DROID setup

For the best performance, we recommend using an SO-100 with the standard wrist configuration and a third-person camera.

For the best performance, please build your Bimanual YAM setup following the reference design below:

All required components can be purchased using this Bimanual YAM parts list.

Implementation code for setting up, data collection, and inference for Bimanual YAM is here

For the Franka setup, we recommend following the official DROID implementation for best results.

Full code for training, fine-tuning, deployment, evaluation, and more details are coming soon.

This model is licensed under Apache 2.0. It is intended for research and educational use in accordance with Ai2's Responsible Use Guidelines.

MolmoAct2 generate robot actions from visual observations and language instructions, but their behavior may vary across embodiments, environments, and hardware configurations. Users should carefully validate model outputs before deployment, especially when operating physical robots or other actuated systems. Where possible, actions should be monitored through interpretable intermediate outputs (adaptive depth map), simulation rollouts, action limits, or other safety checks before execution on hardware. The model’s action space should be bounded by the training data, robot controller limits, and task-specific safety constraints, including limits on speed, workspace, torque, and contact force. Users should follow the hardware manufacturer’s safety guidelines, use appropriate emergency-stop mechanisms, and operate the system only in a safely configured environment with human supervision.

For questions, collaborations, or support, please contact with:

{hqfang,duanj1}@cs.washington.edu 

Found a bug or have a feature request? Please open a GitHub issue.

@misc{fang2026molmoact2actionreasoningmodels,
      title={MolmoAct2: Action Reasoning Models for Real-world Deployment}, 
      author={Haoquan Fang and Jiafei Duan and Donovan Clay and Sam Wang and Shuo Liu and Weikai Huang and Xiang Fan and Wei-Chuan Tsai and Shirui Chen and Yi Ru Wang and Shanli Xing and Jaemin Cho and Jae Sung Park and Ainaz Eftekhar and Peter Sushko and Karen Farley and Angad Wadhwa and Cole Harrison and Winson Han and Ying-Chun Lee and Eli VanderBilt and Rose Hendrix and Suveen Ellawela and Lucas Ngoo and Joyce Chai and Zhongzheng Ren and Ali Farhadi and Dieter Fox and Ranjay Krishna},
      year={2026},
      eprint={2605.02881},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2605.02881}, 
}