We present Vchitect-2.0, a parallel transformer architecture designed to scale up video diffusion models for large-scale text-to-video generation. The overall Vchitect-2.0 system has several key designs. (1) By introducing a novel Multimodal Diffusion Block, our approach achieves consistent alignment between text descriptions and generated video frames, while maintaining temporal coherence across sequences. (2) To overcome memory and computational bottlenecks, we propose a Memory-efficient Training framework that incorporates hybrid parallelism and other memory reduction techniques, enabling efficient training of long video sequences on distributed systems. (3) Additionally, our enhanced data processing pipeline ensures the creation of Vchitect T2V DataVerse, a high-quality million-scale training dataset through rigorous annotation and aesthetic evaluation. Extensive benchmarking demonstrates that Vchitect-2.0 outperforms existing methods in video quality, training efficiency, and scalability, serving as a suitable base for high-fidelity video generation.
@article{arxiv.2501.08453,
title = {Vchitect-2.0: Parallel Transformer for Scaling Up Video Diffusion Models},
author = {Weichen Fan and Chenyang Si and Junhao Song and Zhenyu Yang and Yinan He and Long Zhuo and Ziqi Huang and Ziyue Dong and Jingwen He and Dongwei Pan and Yi Wang and Yuming Jiang and Yaohui Wang and Peng Gao and Xinyuan Chen and Hengjie Li and Dahua Lin and Yu Qiao and Ziwei Liu},
journal= {arXiv preprint arXiv:2501.08453},
year = {2025}
}