Cosmos 3: Omnimodal World Models for Physical AI

Published June 18, 2026 · Category: Robotics

Overview

arXiv:2606.02800v3 Announce Type: replace-cross Abstract: We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-transformers architecture. By supporting highly flexible input-output configurations, Cosmos 3 seamlessly unifies critical modalities for Physical AI -- effectively subsuming vision-language models, video generators, world simulators, and world-action models into a single framework. Our evaluation demonstrates that Cosmos 3 establishes a new state-of-the-art across a diverse suite of understanding and generation tasks, demonstrating omnimodal world models as scalable, general-purpose backbones for embodied agents. Our post-trained Cosmos 3 models were ranked as the best open-source Text-to-Image and Image-to-Video models by Artificial Analysis, and the best policy model by RoboArena at the time the technical report was written. To accelerate open research and deployment in Physical AI, we make our code, model checkpoints, curated synthetic datasets, and evaluation benchmark available under the Linux Foundation's OpenMDW-1.1 License at https://github.com/nvidia/cosmos and https://huggingface.co/collections/nvidia/cosmos3. The project website is available at https://research.nvidia.com/labs/cosmos-lab/cosmos3.

Source

Originally published at arxiv.org.

Related Articles

• Self-Supervised Mask-Aware Transformers for Fault-Tolerant FBG Force Sensing in Minimally Invasive Surgical Robotics
• EffiNav: Fusing Depth and Vision-Language for Efficient Object Goal Navigation
• A Scalable Embodied Intelligence Platform for Seamless Real-to-Sim-to-Real Transfer of Household Mobile Manipulation Tasks