{"ID":2827354,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.16093","arxiv_id":"2512.16093","title":"TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times","abstract":"We introduce TurboDiffusion, a video generation acceleration framework that can speed up end-to-end diffusion generation by 100-200x while maintaining video quality. TurboDiffusion mainly relies on several components for acceleration: (1) Attention acceleration: TurboDiffusion uses low-bit SageAttention and trainable Sparse-Linear Attention (SLA) to speed up attention computation. (2) Step distillation: TurboDiffusion adopts rCM for efficient step distillation. (3) W8A8 quantization: TurboDiffusion quantizes model parameters and activations to 8 bits to accelerate linear layers and compress the model. In addition, TurboDiffusion incorporates several other engineering optimizations. We conduct experiments on the Wan2.2-I2V-14B-720P, Wan2.1-T2V-1.3B-480P, Wan2.1-T2V-14B-720P, and Wan2.1-T2V-14B-480P models. Experimental results show that TurboDiffusion achieves 100-200x speedup for video generation even on a single RTX 5090 GPU, while maintaining comparable video quality. The GitHub repository, which includes model checkpoints and easy-to-use code, is available at https://github.com/thu-ml/TurboDiffusion.","short_abstract":"We introduce TurboDiffusion, a video generation acceleration framework that can speed up end-to-end diffusion generation by 100-200x while maintaining video quality. TurboDiffusion mainly relies on several components for acceleration: (1) Attention acceleration: TurboDiffusion uses low-bit SageAttention and trainable S...","url_abs":"https://arxiv.org/abs/2512.16093","url_pdf":"https://arxiv.org/pdf/2512.16093v1","authors":"[\"Jintao Zhang\",\"Kaiwen Zheng\",\"Kai Jiang\",\"Haoxu Wang\",\"Ion Stoica\",\"Joseph E. Gonzalez\",\"Jianfei Chen\",\"Jun Zhu\"]","published":"2025-12-18T02:21:30Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\",\"cs.LG\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":605798,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2827354,"paper_url":"https://arxiv.org/abs/2512.16093","paper_title":"TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times","repo_url":"https://github.com/thu-ml/TurboDiffusion","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}