# Cosmos 3：面向物理AI的全模态世界模型

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-06-01 08:00
- AIHOT 分数：75
- AIHOT 链接：https://aihot.virxact.com/items/cmpyw41oc03vysli3udi21mie
- 原文链接：https://arxiv.org/abs/2606.02800

## AI 摘要

NVIDIA 发布 Cosmos 3 全模态世界模型家族，基于统一混合 Transformer 架构，联合处理与生成语言、图像、视频、音频和动作序列。该模型将视觉-语言模型、视频生成器、世界模拟器及世界-动作模型整合为单一框架，在多项理解与生成任务上达到新 SOTA。技术报告撰写时，后训练版本被 Artificial Analysis 评为最佳开源文生图与图生视频模型，被 RoboArena 评为最佳策略模型。代码、模型权重、合成数据集及评测基准已开源（Linux Foundation OpenMDW-1.1 许可）。

## 正文

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 https://openmdw.ai/license/1-1/ License at https://github.com/nvidia/cosmos}{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 .