Notable breakthroughs in diffusion modeling have propelled rapid improvements in video generation, yet current foundational model still face critical challenges in simultaneously balancing prompt following, motion plausibility, and visual quality. In this report, we introduce Seedance 1.0, a high-performance and inference-efficient video foundation generation model that integrates several core technical improvements: (i) multi-source data curation augmented with precision and meaningful video captioning, enabling comprehensive learning across diverse scenarios; (ii) an efficient architecture design with proposed training paradigm, which allows for natively supporting multi-shot generation and jointly learning of both text-to-video and image-to-video tasks. (iii) carefully-optimized post-training approaches leveraging fine-grained supervised fine-tuning, and video-specific RLHF with multi-dimensional reward mechanisms for comprehensive performance improvements; (iv) excellent model acceleration achieving ~10x inference speedup through multi-stage distillation strategies and system-level optimizations. Seedance 1.0 can generate a 5-second video at 1080p resolution only with 41.4 seconds (NVIDIA-L20). Compared to state-of-the-art video generation models, Seedance 1.0 stands out with high-quality and fast video generation having superior spatiotemporal fluidity with structural stability, precise instruction adherence in complex multi-subject contexts, native multi-shot narrative coherence with consistent subject representation.
View on arXiv@article{gao2025_2506.09113,
title={ Seedance 1.0: Exploring the Boundaries of Video Generation Models },
author={ Yu Gao and Haoyuan Guo and Tuyen Hoang and Weilin Huang and Lu Jiang and Fangyuan Kong and Huixia Li and Jiashi Li and Liang Li and Xiaojie Li and Xunsong Li and Yifu Li and Shanchuan Lin and Zhijie Lin and Jiawei Liu and Shu Liu and Xiaonan Nie and Zhiwu Qing and Yuxi Ren and Li Sun and Zhi Tian and Rui Wang and Sen Wang and Guoqiang Wei and Guohong Wu and Jie Wu and Ruiqi Xia and Fei Xiao and Xuefeng Xiao and Jiangqiao Yan and Ceyuan Yang and Jianchao Yang and Runkai Yang and Tao Yang and Yihang Yang and Zilyu Ye and Xuejiao Zeng and Yan Zeng and Heng Zhang and Yang Zhao and Xiaozheng Zheng and Peihao Zhu and Jiaxin Zou and Feilong Zuo },
journal={arXiv preprint arXiv:2506.09113},
year={ 2025 }
}