PipeFusion: Patch-level Pipeline Parallelism for Diffusion Transformers Inference

Jiarui Fang, Jinzhe Pan, Aoyu Li, Xibo Sun, Jiannan Wang. PipeFusion: Patch-level Pipeline Parallelism for Diffusion Transformers Inference. In Danielle Belgrave, Cheng Zhang 0005, Laura N. Montoya, Hsuan-Tien Lin, Razvan Pascanu, Piotr Koniusz, Marzyeh Ghassemi, Nancy Chen, Iván Vladimir Meza Ruíz, Arturo Loaiza-Bonilla, editors, Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2025, NeurIPS 2025, San Diago, CA, USA, December 2-7, 2025 / Mexico City, Mexico, November 30 - December 5, 2025. 2025. [doi]

Authors

Jiarui Fang

This author has not been identified. Look up 'Jiarui Fang' in Google

Jinzhe Pan

This author has not been identified. Look up 'Jinzhe Pan' in Google

Aoyu Li

This author has not been identified. Look up 'Aoyu Li' in Google

Xibo Sun

This author has not been identified. Look up 'Xibo Sun' in Google

Jiannan Wang

This author has not been identified. Look up 'Jiannan Wang' in Google