OpenWAM

AI model
Humanoids and robot learning

OpenWAM is an open, modular research stack for developing World-Action Models, pairing infrastructure for composing model and training choices with controlled studies and a pretrained model.

Made by OpenWAM.

Source on GitHub: github.com/OpenWAM-Official/OpenWAM (opens in a new tab). Facts read at commit 1e103e1 (opens in a new tab).

Licenses

Code: Apache-2.0 (opens in a new tab)
Read by Robotheca from LICENSE (opens in a new tab) (Apache-2.0) at commit 1e103e1 (opens in a new tab), on 16 Oct 2026.
Weights: Apache-2.0 (opens in a new tab)
Read on huggingface.co/OpenWAM/OpenWAM-Alpha-Real-Dexterous-Hand-Wu… (opens in a new tab).
Commercial use: Yes

What it does

Modular infrastructure for composing and comparing model, representation, training, inference, deployment and evaluation choices.
Controlled studies deriving practical principles for inheriting world knowledge, coupling world and action learning, and scaling across domains.
OpenWAM-α, an open pretrained World-Action Model trained on 518.5M frames of egocentric human and robot data.
Architectures are selected through Hydra YAML configs, with several variants such as dual_system and single_system.
Supports benchmark evaluation on RoboTwin, LIBERO, RoboCasa, VLABench and EBench, plus fine-tuning and WebSocket policy deployment.

Written by Robotheca from the project's README (opens in a new tab) at commit 1e103e1.

Getting started

Difficulty: Advanced
How to start: The project's website (opens in a new tab)

Maintenance

Status: Active: a commit in the last 90 days
Newest commit: 14 Oct 2026 on main
Latest release: None published on GitHub

Checked on GitHub on 16 Oct 2026.

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.