Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model
Dynin-Robotics unifies action, goal, and dynamics prediction in one omnimodal masked-diffusion VLA model, reaching 78.4% success on Franka Research 3 manipulation tasks.
Built on the Dynin-Omni masked-diffusion backbone, the model represents language, observations, goals, and actions as discrete tokens and is continually pretrained on roughly 1.33 million trajectories from 48 Open X-Embodiment datasets. The shared trajectory interface enables test-time scaling via goal prediction, action-candidate evaluation, and joint action/future-state refinement. It achieves competitive results on LIBERO and zero-shot LIBERO-Plus, 78.4% average success across four Franka Research 3 conditions, and up to 29.2x faster model-side action decoding from a block-parallel implementation.
- Single omnimodal model learns action prediction, next-observation prediction, goal prediction, and instruction reconstruction.
- Continually pretrained on ~1.33M trajectories from 48 Open X-Embodiment datasets.
- 78.4% average success across four Franka Research 3 manipulation conditions.
- Block-parallel implementation speeds model-side action decoding up to 29.2x.
From the original article
1 min read at huggingface.co
Visual goal and dynamics prediction can provide language-conditioned robot policies with both a target outcome and a representation of action-dependent scene changes. We bring these predictions into action generation and selection through a shared trajectory model.
By Hoeun Lee, Jaeik Kim, Jusang Oh · 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.