Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.25180 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908570714374144 |
|---|---|
| author | He, Wenkun Gu, Yuchao Chen, Junyu Zou, Dongyun Lin, Yujun Zhang, Zhekai Xi, Haocheng Li, Muyang Zhu, Ligeng Yu, Jincheng Chen, Junsong Xie, Enze Han, Song Cai, Han |
| author_facet | He, Wenkun Gu, Yuchao Chen, Junyu Zou, Dongyun Lin, Yujun Zhang, Zhekai Xi, Haocheng Li, Muyang Zhu, Ligeng Yu, Jincheng Chen, Junsong Xie, Enze Han, Song Cai, Han |
| contents | Existing text-to-image diffusion models excel at generating high-quality images, but face significant efficiency challenges when scaled to high resolutions, like 4K image generation. While previous research accelerates diffusion models in various aspects, it seldom handles the inherent redundancy within the latent space. To bridge this gap, this paper introduces DC-Gen, a general framework that accelerates text-to-image diffusion models by leveraging a deeply compressed latent space. Rather than a costly training-from-scratch approach, DC-Gen uses an efficient post-training pipeline to preserve the quality of the base model. A key challenge in this paradigm is the representation gap between the base model's latent space and a deeply compressed latent space, which can lead to instability during direct fine-tuning. To overcome this, DC-Gen first bridges the representation gap with a lightweight embedding alignment training. Once the latent embeddings are aligned, only a small amount of LoRA fine-tuning is needed to unlock the base model's inherent generation quality. We verify DC-Gen's effectiveness on SANA and FLUX.1-Krea. The resulting DC-Gen-SANA and DC-Gen-FLUX models achieve quality comparable to their base models but with a significant speedup. Specifically, DC-Gen-FLUX reduces the latency of 4K image generation by 53x on the NVIDIA H100 GPU. When combined with NVFP4 SVDQuant, DC-Gen-FLUX generates a 4K image in just 3.5 seconds on a single NVIDIA 5090 GPU, achieving a total latency reduction of 138x compared to the base FLUX.1-Krea model. Code: https://github.com/dc-ai-projects/DC-Gen. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25180 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space He, Wenkun Gu, Yuchao Chen, Junyu Zou, Dongyun Lin, Yujun Zhang, Zhekai Xi, Haocheng Li, Muyang Zhu, Ligeng Yu, Jincheng Chen, Junsong Xie, Enze Han, Song Cai, Han Computer Vision and Pattern Recognition Artificial Intelligence Existing text-to-image diffusion models excel at generating high-quality images, but face significant efficiency challenges when scaled to high resolutions, like 4K image generation. While previous research accelerates diffusion models in various aspects, it seldom handles the inherent redundancy within the latent space. To bridge this gap, this paper introduces DC-Gen, a general framework that accelerates text-to-image diffusion models by leveraging a deeply compressed latent space. Rather than a costly training-from-scratch approach, DC-Gen uses an efficient post-training pipeline to preserve the quality of the base model. A key challenge in this paradigm is the representation gap between the base model's latent space and a deeply compressed latent space, which can lead to instability during direct fine-tuning. To overcome this, DC-Gen first bridges the representation gap with a lightweight embedding alignment training. Once the latent embeddings are aligned, only a small amount of LoRA fine-tuning is needed to unlock the base model's inherent generation quality. We verify DC-Gen's effectiveness on SANA and FLUX.1-Krea. The resulting DC-Gen-SANA and DC-Gen-FLUX models achieve quality comparable to their base models but with a significant speedup. Specifically, DC-Gen-FLUX reduces the latency of 4K image generation by 53x on the NVIDIA H100 GPU. When combined with NVFP4 SVDQuant, DC-Gen-FLUX generates a 4K image in just 3.5 seconds on a single NVIDIA 5090 GPU, achieving a total latency reduction of 138x compared to the base FLUX.1-Krea model. Code: https://github.com/dc-ai-projects/DC-Gen. |
| title | DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2509.25180 |