_version_ 1866908704138330112
author Seedream, Team
:
Chen, Yunpeng
Gao, Yu
Gong, Lixue
Guo, Meng
Guo, Qiushan
Guo, Zhiyao
Hou, Xiaoxia
Huang, Weilin
Huang, Yixuan
Jian, Xiaowen
Kuang, Huafeng
Lai, Zhichao
Li, Fanshi
Li, Liang
Lian, Xiaochen
Liao, Chao
Liu, Liyang
Liu, Wei
Lu, Yanzuo
Luo, Zhengxiong
Ou, Tongtong
Shi, Guang
Shi, Yichun
Sun, Shiqi
Tian, Yu
Tian, Zhi
Wang, Peng
Wang, Rui
Wang, Xun
Wang, Ye
Wu, Guofeng
Wu, Jie
Wu, Wenxu
Wu, Yonghui
Xia, Xin
Xiao, Xuefeng
Xu, Shuang
Yan, Xin
Yang, Ceyuan
Yang, Jianchao
Zhai, Zhonghua
Zhang, Chenlin
Zhang, Heng
Zhang, Qi
Zhang, Xinyu
Zhang, Yuwei
Zhao, Shijia
Zhao, Wenliang
Zhu, Wenjia
author_facet Seedream, Team
:
Chen, Yunpeng
Gao, Yu
Gong, Lixue
Guo, Meng
Guo, Qiushan
Guo, Zhiyao
Hou, Xiaoxia
Huang, Weilin
Huang, Yixuan
Jian, Xiaowen
Kuang, Huafeng
Lai, Zhichao
Li, Fanshi
Li, Liang
Lian, Xiaochen
Liao, Chao
Liu, Liyang
Liu, Wei
Lu, Yanzuo
Luo, Zhengxiong
Ou, Tongtong
Shi, Guang
Shi, Yichun
Sun, Shiqi
Tian, Yu
Tian, Zhi
Wang, Peng
Wang, Rui
Wang, Xun
Wang, Ye
Wu, Guofeng
Wu, Jie
Wu, Wenxu
Wu, Yonghui
Xia, Xin
Xiao, Xuefeng
Xu, Shuang
Yan, Xin
Yang, Ceyuan
Yang, Jianchao
Zhai, Zhonghua
Zhang, Chenlin
Zhang, Heng
Zhang, Qi
Zhang, Xinyu
Zhang, Yuwei
Zhao, Shijia
Zhao, Wenliang
Zhu, Wenjia
contents We introduce Seedream 4.0, an efficient and high-performance multimodal image generation system that unifies text-to-image (T2I) synthesis, image editing, and multi-image composition within a single framework. We develop a highly efficient diffusion transformer with a powerful VAE which also can reduce the number of image tokens considerably. This allows for efficient training of our model, and enables it to fast generate native high-resolution images (e.g., 1K-4K). Seedream 4.0 is pretrained on billions of text-image pairs spanning diverse taxonomies and knowledge-centric concepts. Comprehensive data collection across hundreds of vertical scenarios, coupled with optimized strategies, ensures stable and large-scale training, with strong generalization. By incorporating a carefully fine-tuned VLM model, we perform multi-modal post-training for training both T2I and image editing tasks jointly. For inference acceleration, we integrate adversarial distillation, distribution matching, and quantization, as well as speculative decoding. It achieves an inference time of up to 1.8 seconds for generating a 2K image (without a LLM/VLM as PE model). Comprehensive evaluations reveal that Seedream 4.0 can achieve state-of-the-art results on both T2I and multimodal image editing. In particular, it demonstrates exceptional multimodal capabilities in complex tasks, including precise image editing and in-context reasoning, and also allows for multi-image reference, and can generate multiple output images. This extends traditional T2I systems into an more interactive and multidimensional creative tool, pushing the boundary of generative AI for both creativity and professional applications. We further scale our model and data as Seedream 4.5. Seedream 4.0 and Seedream 4.5 are accessible on Volcano Engine https://www.volcengine.com/experience/ark?launch=seedream.
format Preprint
id arxiv_https___arxiv_org_abs_2509_20427
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Seedream 4.0: Toward Next-generation Multimodal Image Generation
Seedream, Team
:
Chen, Yunpeng
Gao, Yu
Gong, Lixue
Guo, Meng
Guo, Qiushan
Guo, Zhiyao
Hou, Xiaoxia
Huang, Weilin
Huang, Yixuan
Jian, Xiaowen
Kuang, Huafeng
Lai, Zhichao
Li, Fanshi
Li, Liang
Lian, Xiaochen
Liao, Chao
Liu, Liyang
Liu, Wei
Lu, Yanzuo
Luo, Zhengxiong
Ou, Tongtong
Shi, Guang
Shi, Yichun
Sun, Shiqi
Tian, Yu
Tian, Zhi
Wang, Peng
Wang, Rui
Wang, Xun
Wang, Ye
Wu, Guofeng
Wu, Jie
Wu, Wenxu
Wu, Yonghui
Xia, Xin
Xiao, Xuefeng
Xu, Shuang
Yan, Xin
Yang, Ceyuan
Yang, Jianchao
Zhai, Zhonghua
Zhang, Chenlin
Zhang, Heng
Zhang, Qi
Zhang, Xinyu
Zhang, Yuwei
Zhao, Shijia
Zhao, Wenliang
Zhu, Wenjia
Computer Vision and Pattern Recognition
We introduce Seedream 4.0, an efficient and high-performance multimodal image generation system that unifies text-to-image (T2I) synthesis, image editing, and multi-image composition within a single framework. We develop a highly efficient diffusion transformer with a powerful VAE which also can reduce the number of image tokens considerably. This allows for efficient training of our model, and enables it to fast generate native high-resolution images (e.g., 1K-4K). Seedream 4.0 is pretrained on billions of text-image pairs spanning diverse taxonomies and knowledge-centric concepts. Comprehensive data collection across hundreds of vertical scenarios, coupled with optimized strategies, ensures stable and large-scale training, with strong generalization. By incorporating a carefully fine-tuned VLM model, we perform multi-modal post-training for training both T2I and image editing tasks jointly. For inference acceleration, we integrate adversarial distillation, distribution matching, and quantization, as well as speculative decoding. It achieves an inference time of up to 1.8 seconds for generating a 2K image (without a LLM/VLM as PE model). Comprehensive evaluations reveal that Seedream 4.0 can achieve state-of-the-art results on both T2I and multimodal image editing. In particular, it demonstrates exceptional multimodal capabilities in complex tasks, including precise image editing and in-context reasoning, and also allows for multi-image reference, and can generate multiple output images. This extends traditional T2I systems into an more interactive and multidimensional creative tool, pushing the boundary of generative AI for both creativity and professional applications. We further scale our model and data as Seedream 4.5. Seedream 4.0 and Seedream 4.5 are accessible on Volcano Engine https://www.volcengine.com/experience/ark?launch=seedream.
title Seedream 4.0: Toward Next-generation Multimodal Image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.20427