A Modular Open-Source Framework for Systematic Training, Inference, and Deployment of World-Action Models
📦 Huggingface Collections
🔬 OpenWAM-Study: Public Release of checkpoints from our controlled study on WAM design principles.
🚀 OpenWAM-Alpha: Public Release of our pretrained checkpoint (OpenWAM-Alpha), plus finetuned variants across downstream benchmarks.