HunyuanVideo: A Systematic Framework For Large Video Generative Models

TL;DR

HunyuanVideo is an open-source large-scale video generative model that achieves state-of-the-art performance, surpassing many existing models, and aims to democratize access to advanced video generation technology.

Contribution

The paper introduces HunyuanVideo, the largest open-source video foundation model with over 13 billion parameters, and presents a comprehensive framework for high-quality, large-scale video generation.

Findings

HunyuanVideo outperforms previous state-of-the-art models.

The model demonstrates high visual quality and accurate motion dynamics.

Open-source release fosters community experimentation and innovation.

Abstract

Recent advancements in video generation have significantly impacted daily life for both individuals and industries. However, the leading video generation models remain closed-source, resulting in a notable performance gap between industry capabilities and those available to the public. In this report, we introduce HunyuanVideo, an innovative open-source video foundation model that demonstrates performance in video generation comparable to, or even surpassing, that of leading closed-source models. HunyuanVideo encompasses a comprehensive framework that integrates several key elements, including data curation, advanced architectural design, progressive model scaling and training, and an efficient infrastructure tailored for large-scale model training and inference. As a result, we successfully trained a video generative model with over 13 billion parameters, making it the largest among all open-source models. We conducted extensive experiments and implemented a series of targeted designs to ensure high visual quality, motion dynamics, text-video alignment, and advanced filming techniques. According to evaluations by professionals, HunyuanVideo outperforms previous state-of-the-art models, including Runway Gen-3, Luma 1.6, and three top-performing Chinese video generative models. By releasing the code for the foundation model and its applications, we aim to bridge the gap between closed-source and open-source communities. This initiative will empower individuals within the community to experiment with their ideas, fostering a more dynamic and vibrant video generation ecosystem. The code is publicly available at https://github.com/Tencent/HunyuanVideo.