MolmoAct: Action Reasoning Models that can Reason in Space
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909795698606080 |
|---|---|
| author | Lee, Jason Duan, Jiafei Fang, Haoquan Deng, Yuquan Liu, Shuo Li, Boyang Fang, Bohan Zhang, Jieyu Wang, Yi Ru Lee, Sangho Han, Winson Pumacay, Wilbert Wu, Angelica Hendrix, Rose Farley, Karen VanderBilt, Eli Farhadi, Ali Fox, Dieter Krishna, Ranjay |
| author_facet | Lee, Jason Duan, Jiafei Fang, Haoquan Deng, Yuquan Liu, Shuo Li, Boyang Fang, Bohan Zhang, Jieyu Wang, Yi Ru Lee, Sangho Han, Winson Pumacay, Wilbert Wu, Angelica Hendrix, Rose Farley, Karen VanderBilt, Eli Farhadi, Ali Fox, Dieter Krishna, Ranjay |
| contents | Reasoning is central to purposeful action, yet most robotic foundation models map perception and instructions directly to control, which limits adaptability, generalization, and semantic grounding. We introduce Action Reasoning Models (ARMs), a class of robotic foundation models that integrate perception, planning, and control through a structured three-stage pipeline. Our model, MolmoAct, encodes observations and instructions into depth-aware perception tokens, generates mid-level spatial plans as editable trajectory traces, and predicts precise low-level actions, enabling explainable and steerable behavior. MolmoAct-7B-D achieves strong performance across simulation and real-world settings: 70.5% zero-shot accuracy on SimplerEnv Visual Matching tasks, surpassing closed-source Pi-0 and GR00T N1.5; 86.6% average success on LIBERO, including an additional 6.3% gain over ThinkAct on long-horizon tasks; and in real-world fine-tuning, an additional 10% (single-arm) and an additional 22.7% (bimanual) task progression over Pi-0-FAST. It also outperforms baselines by an additional 23.3% on out-of-distribution generalization and achieves top human-preference scores for open-ended instruction following and trajectory steering. Furthermore, we release, for the first time, the MolmoAct Dataset -- a mid-training robot dataset comprising over 10,000 high quality robot trajectories across diverse scenarios and tasks. Training with this dataset yields an average 5.5% improvement in general performance over the base model. We release all model weights, training code, our collected dataset, and our action reasoning dataset, establishing MolmoAct as both a state-of-the-art robotics foundation model and an open blueprint for building ARMs that transform perception into purposeful action through structured reasoning. Blogpost: https://allenai.org/blog/molmoact |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_07917 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MolmoAct: Action Reasoning Models that can Reason in Space Lee, Jason Duan, Jiafei Fang, Haoquan Deng, Yuquan Liu, Shuo Li, Boyang Fang, Bohan Zhang, Jieyu Wang, Yi Ru Lee, Sangho Han, Winson Pumacay, Wilbert Wu, Angelica Hendrix, Rose Farley, Karen VanderBilt, Eli Farhadi, Ali Fox, Dieter Krishna, Ranjay Robotics Reasoning is central to purposeful action, yet most robotic foundation models map perception and instructions directly to control, which limits adaptability, generalization, and semantic grounding. We introduce Action Reasoning Models (ARMs), a class of robotic foundation models that integrate perception, planning, and control through a structured three-stage pipeline. Our model, MolmoAct, encodes observations and instructions into depth-aware perception tokens, generates mid-level spatial plans as editable trajectory traces, and predicts precise low-level actions, enabling explainable and steerable behavior. MolmoAct-7B-D achieves strong performance across simulation and real-world settings: 70.5% zero-shot accuracy on SimplerEnv Visual Matching tasks, surpassing closed-source Pi-0 and GR00T N1.5; 86.6% average success on LIBERO, including an additional 6.3% gain over ThinkAct on long-horizon tasks; and in real-world fine-tuning, an additional 10% (single-arm) and an additional 22.7% (bimanual) task progression over Pi-0-FAST. It also outperforms baselines by an additional 23.3% on out-of-distribution generalization and achieves top human-preference scores for open-ended instruction following and trajectory steering. Furthermore, we release, for the first time, the MolmoAct Dataset -- a mid-training robot dataset comprising over 10,000 high quality robot trajectories across diverse scenarios and tasks. Training with this dataset yields an average 5.5% improvement in general performance over the base model. We release all model weights, training code, our collected dataset, and our action reasoning dataset, establishing MolmoAct as both a state-of-the-art robotics foundation model and an open blueprint for building ARMs that transform perception into purposeful action through structured reasoning. Blogpost: https://allenai.org/blog/molmoact |
| title | MolmoAct: Action Reasoning Models that can Reason in Space |
| topic | Robotics |
| url | https://arxiv.org/abs/2508.07917 |