MolmoAct
- AI model
- Robot arms and industry
A 7B vision-language-action model from Ai2 that reasons in space before it acts: it predicts a depth map and the end effector's path in the image, then the action. Built on Qwen2.5-7B with a SigLIP2 vision encoder.
- Non-commercial
Made by Allen Institute for AI (Ai2).
Source on Other: huggingface.co/allenai/MolmoAct-7B-D-0812 (opens in a new tab).
Licenses
- Code
- Apache-2.0 (opens in a new tab)
A person at Robotheca set it on 23 Oct 2026. Checked by hand on 23 Oct 2026.
- Weights
- Apache-2.0 (opens in a new tab)
Read on huggingface.co/allenai/MolmoAct-7B-D-0812 (opens in a new tab).
- Commercial use
- No
The assets license does not allow commercial use.
Labels
- Non-commercialThe assets license (CC-BY-NC-4.0) does not allow commercial use.
What it does
Robotheca has not written what it does yet. Until then, see the project's own page: huggingface.co/allenai/MolmoAct-7B-D-0812 (opens in a new tab).
Getting started
- Difficulty
- Advanced
Maintenance
- Status
- Not checked yet
Checked by hand on 23 Oct 2026.
Links
Robots that can run it
No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.
Robotheca lists AI models as a reference: nothing is hosted or run here, and every fact links to where it was read.