OpenWAM: An Open, Modular Exploration Towards Systematic World-Action Model Pretraining
OpenWAM releases an open modular stack for world-action model pretraining, plus OpenWAM-alpha trained on about 6,400 hours of egocentric and robot data.
OpenWAM is an open research stack that factorizes World-Action Model pretraining into composable infrastructure, study, and model components with unified training, inference, and evaluation. Controlled experiments distill three principles on knowledge inheritance, world-action synergy, and out-of-domain generalization gains from embodied co-training. The resulting OpenWAM-alpha, pretrained on roughly 6,400 hours of egocentric human and robot data, achieves top-tier results across eight simulation benchmarks and real-robot tests spanning single-arm, bimanual, and dexterous embodiments. The full stack, including infrastructure, evaluation protocols, pretrained models, and data recipes, is released openly.
- OpenWAM-Infra decomposes world-action pretraining into composable modules
- Controlled studies identify what to inherit and how world-action learning interacts
- Principles: capable generative backbone, dedicated action capacity, joint denoising
- OpenWAM-alpha trained on ~6,400 hours of egocentric and robot data
- Strong simulation-to-real transfer across arm, bimanual, and dexterous setups
- Tags
- embodied-aimanipulationopen-sourceroboticssim-to-realvideo-generation-pretrainingworld-action-model
- AI models
- OpenWAM-alpha
From the original article
1 min read at huggingface.co
World-Action Models inherit world knowledge from video-generative priors, and channel it into executable control signals through embodied experience. Existing systems, however, are monolithic: the generative backbone, visual representation, architecture, information flow, inference procedure, and training data are tightly coupled, obscuring which design choices matter and why.
By Yuran Wang, Siqiao Huang, Mingleyang Li · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.