MolmoB0T: Large-Scale Simulation Enables Zero-Shot Manipulation

Authors: Abhay Deshpande, Maya Guru, Rose Hendrix, Snehal Jauhri, Ainaz Eftekhar, Rohun Tripathi, Max Argus, Jordi Salvador, Haoquan Fang, Matthew Wallingford, Wilbert Pumacay, Yejin Kim, Quinn Pfeifer, Ying-Chun Lee, Piper Wolters, Omar Rayyan, Mingtong Zhang, Jiafei Duan, Karen Farley, Winson Han, Eli Vanderbilt, Dieter Fox, Ali Farhadi, Georgia Chalvatzaki, Dhruv Shah, Ranjay Krishna · CoRL 2026

Licences

• Open Access
• Paper: arXiv.org perpetual, non-exclusive license 1.0

Under arXiv's licence, everything below is written from the abstract only.

Why it matters

Sim-to-real transfer is widely believed to need some real-world data or task-specific fine-tuning. The authors report zero-shot transfer from simulation alone to unseen objects and environments, with MolmoBot reaching 79.2% success on real tabletop pick-and-place across four settings against 39.2% for π0.5.

Proposed solution

MolmoBot-Engine is an open-source pipeline that procedurally generates data across robots, tasks and simulated environments in MolmoSpaces; the authors release MolmoBot-Data (1.7 million expert trajectories) and train three policy classes, evaluated on a Franka FR3 for tabletop tasks and a Rainbow Robotics RB-Y1 mobile manipulator.

Abstract

A prevailing view in robot learning is that simulation alone is not enough; effective sim-to-real transfer is widely believed to require at least some real-world data collection or task-specific fine-tuning to bridge the gap between simulated and physical environments. We challenge that assumption. With sufficiently large-scale and diverse simulated synthetic training data, we show that zero-shot transfer to the real world is not only possible, but effective for both static and mobile manipulation. We introduce MolmoBot-Engine, a fully open-source pipeline for procedural data generation across robots, tasks, and diverse simulated environments in MolmoSpaces. With it, we release MolmoBot-Data, a dataset of 1.7 million expert trajectories for articulated object manipulation and pick-and-place tasks. We train three policy classes: MolmoBot, a Molmo2-based multi-frame vision-language model with a flow-matching action head; MolmoBot-Pi0, which replicates the π0 architecture to enable direct comparison; and MolmoBot-SPOC, a lightweight policy suitable for edge deployment and amenable to RL fine-tuning. We evaluate on two robotic platforms: the Franka FR3 for tabletop manipulation tasks and the Rainbow Robotics RB-Y1 mobile manipulator for door opening, drawer manipulation, cabinet interaction, and mobile pick-and-place. Without any real-world fine-tuning, our policies achieve zero-shot transfer to unseen objects and environments. On tabletop pick-and-place, MolmoBot achieves a success rate of 79.2% in real world evaluations across 4 settings, outperforming π0.5 at 39.2%. Our results demonstrate that procedural environment generation combined with diverse articulated assets can produce robust manipulation policies that generalize broadly to the real world.

Links

• arXiv 2603.16861: https://arxiv.org/abs/2603.16861
• PDF: https://arxiv.org/pdf/2603.16861
• Code: https://github.com/allenai/MolmoBot
• Project page: https://allenai.github.io/MolmoBot/
• Dataset: https://huggingface.co/datasets/allenai/molmobot-data