Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917132768378880 |
|---|---|
| author | Image Team Cai, Huanqia Cao, Sihan Du, Ruoyi Gao, Peng Hoi, Steven Hou, Zhaohui Huang, Shijie Jiang, Dengyang Jin, Xin Li, Liangchen Li, Zhen Li, Zhong-Yu Liu, David Liu, Dongyang Shi, Junhan Wu, Qilong Yu, Feng Zhang, Chi Zhang, Shifeng Zhou, Shilin |
| author_facet | Image Team Cai, Huanqia Cao, Sihan Du, Ruoyi Gao, Peng Hoi, Steven Hou, Zhaohui Huang, Shijie Jiang, Dengyang Jin, Xin Li, Liangchen Li, Zhen Li, Zhong-Yu Liu, David Liu, Dongyang Shi, Junhan Wu, Qilong Yu, Feng Zhang, Chi Zhang, Shifeng Zhou, Shilin |
| contents | The landscape of high-performance image generation models is currently dominated by proprietary systems, such as Nano Banana Pro and Seedream 4.0. Leading open-source alternatives, including Qwen-Image, Hunyuan-Image-3.0 and FLUX.2, are characterized by massive parameter counts (20B to 80B), making them impractical for inference, and fine-tuning on consumer-grade hardware. To address this gap, we propose Z-Image, an efficient 6B-parameter foundation generative model built upon a Scalable Single-Stream Diffusion Transformer (S3-DiT) architecture that challenges the "scale-at-all-costs" paradigm. By systematically optimizing the entire model lifecycle -- from a curated data infrastructure to a streamlined training curriculum -- we complete the full training workflow in just 314K H800 GPU hours (approx. $630K). Our few-step distillation scheme with reward post-training further yields Z-Image-Turbo, offering both sub-second inference latency on an enterprise-grade H800 GPU and compatibility with consumer-grade hardware (<16GB VRAM). Additionally, our omni-pre-training paradigm also enables efficient training of Z-Image-Edit, an editing model with impressive instruction-following capabilities. Both qualitative and quantitative experiments demonstrate that our model achieves performance comparable to or surpassing that of leading competitors across various dimensions. Most notably, Z-Image exhibits exceptional capabilities in photorealistic image generation and bilingual text rendering, delivering results that rival top-tier commercial models, thereby demonstrating that state-of-the-art results are achievable with significantly reduced computational overhead. We publicly release our code, weights, and online demo to foster the development of accessible, budget-friendly, yet state-of-the-art generative models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_22699 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer Image Team Cai, Huanqia Cao, Sihan Du, Ruoyi Gao, Peng Hoi, Steven Hou, Zhaohui Huang, Shijie Jiang, Dengyang Jin, Xin Li, Liangchen Li, Zhen Li, Zhong-Yu Liu, David Liu, Dongyang Shi, Junhan Wu, Qilong Yu, Feng Zhang, Chi Zhang, Shifeng Zhou, Shilin Computer Vision and Pattern Recognition The landscape of high-performance image generation models is currently dominated by proprietary systems, such as Nano Banana Pro and Seedream 4.0. Leading open-source alternatives, including Qwen-Image, Hunyuan-Image-3.0 and FLUX.2, are characterized by massive parameter counts (20B to 80B), making them impractical for inference, and fine-tuning on consumer-grade hardware. To address this gap, we propose Z-Image, an efficient 6B-parameter foundation generative model built upon a Scalable Single-Stream Diffusion Transformer (S3-DiT) architecture that challenges the "scale-at-all-costs" paradigm. By systematically optimizing the entire model lifecycle -- from a curated data infrastructure to a streamlined training curriculum -- we complete the full training workflow in just 314K H800 GPU hours (approx. $630K). Our few-step distillation scheme with reward post-training further yields Z-Image-Turbo, offering both sub-second inference latency on an enterprise-grade H800 GPU and compatibility with consumer-grade hardware (<16GB VRAM). Additionally, our omni-pre-training paradigm also enables efficient training of Z-Image-Edit, an editing model with impressive instruction-following capabilities. Both qualitative and quantitative experiments demonstrate that our model achieves performance comparable to or surpassing that of leading competitors across various dimensions. Most notably, Z-Image exhibits exceptional capabilities in photorealistic image generation and bilingual text rendering, delivering results that rival top-tier commercial models, thereby demonstrating that state-of-the-art results are achievable with significantly reduced computational overhead. We publicly release our code, weights, and online demo to foster the development of accessible, budget-friendly, yet state-of-the-art generative models. |
| title | Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.22699 |