Year: 2025
Type: preprint
Abstract: We present Seed Diffusion Preview, a large-scale language model based on discrete-state diffusion, offering remarkably fast inference speed. Thanks to non-sequential, parallel generation, discrete diffusion models provide a notable speedup to mitigate the inherent latency of token-by-token decoding, as demonstrated recently (e.g., Mercury Coder, Gemini Diffusion). Seed Diffusion Preview achieves an inference speed of 2,146 token/s over H20 GPUs while maintaining competitive performance across a sweep of standard code evaluation benchmarks, significantly faster than contemporary Mercury and Gemini Diffusion, establishing new state of the art on the speed-quality Pareto frontier for code models.
Source: arXiv (Cornell University)
Authors: Song, Yuxuan, Zhang, Zheng, Luo, Cheng, Gao, Pengyang, Xia, Fan, Luo, Hao, Li, Zheng, Yang, Yuehang, Yu, Hongli, Qu, Xingwei, Fu, Yuwei, Su, Jing, Zhang, Ge, Huang, Wenhao, Wang, Mingxuan, Yan, Lin, Jia, Xiaoying, Liu, Jingjing, Ma, Wei-Ying, Zhang, Ya-Qin, Wu, Yonghui, Zhou, Hao
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4417102600