StreamRL: Scalable, Heterogeneous, and Elastic RL for LLMs with Disaggregated Stream Generation

no code implementations • 22 Apr 2025 • Yinmin Zhong, Zili Zhang, Xiaoniu Song, Hanpeng Hu, Chao Jin, Bingyang Wu, Nuo Chen, Yukun Chen, Yu Zhou, Changyi Wan, HongYu Zhou, Yimin Jiang, Yibo Zhu, Daxin Jiang

However, in real-world deployments, we observe that the colocated architecture suffers from resource coupling, where the two stages are constrained to use the same resources.

Reinforcement Learning (RL) Scheduling

Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction

1 code implementation • 17 Feb 2025 • Ailin Huang, Boyong Wu, Bruce Wang, Chao Yan, Chen Hu, Chengli Feng, Fei Tian, Feiyu Shen, Jingbei Li, Mingrui Chen, Peng Liu, Ruihang Miao, Wang You, Xi Chen, Xuerui Yang, Yechang Huang, Yuxiang Zhang, Zheng Gong, Zixin Zhang, HongYu Zhou, Jianjian Sun, Brian Li, Chengting Feng, Changyi Wan, Hanpeng Hu, Jianchang Wu, Jiangjie Zhen, Ranchen Ming, Song Yuan, Xuelin Zhang, Yu Zhou, Bingxin Li, Buyun Ma, Hongyuan Wang, Kang An, Wei Ji, Wen Li, Xuan Wen, Xiangwen Kong, Yuankai Ma, Yuanwei Liang, Yun Mou, Bahtiyar Ahmidi, Bin Wang, Bo Li, Changxin Miao, Chen Xu, Chenrun Wang, Dapeng Shi, Deshan Sun, Dingyuan Hu, Dula Sai, Enle Liu, Guanzhe Huang, Gulin Yan, Heng Wang, Haonan Jia, Haoyang Zhang, Jiahao Gong, Junjing Guo, Jiashuai Liu, Jiahong Liu, Jie Feng, Jie Wu, Jiaoren Wu, Jie Yang, Jinguo Wang, Jingyang Zhang, Junzhe Lin, Kaixiang Li, Lei Xia, Li Zhou, Liang Zhao, Longlong Gu, Mei Chen, Menglin Wu, Ming Li, Mingxiao Li, Mingliang Li, Mingyao Liang, Na Wang, Nie Hao, Qiling Wu, Qinyuan Tan, Ran Sun, Shuai Shuai, Shaoliang Pang, Shiliang Yang, Shuli Gao, Shanshan Yuan, SiQi Liu, Shihong Deng, Shilei Jiang, Sitong Liu, Tiancheng Cao, Tianyu Wang, Wenjin Deng, Wuxun Xie, Weipeng Ming, Wenqing He, Wen Sun, Xin Han, Xin Huang, Xiaomin Deng, Xiaojia Liu, Xin Wu, Xu Zhao, Yanan Wei, Yanbo Yu, Yang Cao, Yangguang Li, Yangzhen Ma, Yanming Xu, Yaoyu Wang, Yaqiang Shi, Yilei Wang, Yizhuang Zhou, Yinmin Zhong, Yang Zhang, Yaoben Wei, Yu Luo, Yuanwei Lu, Yuhe Yin, Yuchu Luo, Yuanhao Ding, Yuting Yan, Yaqi Dai, Yuxiang Yang, Zhe Xie, Zheng Ge, Zheng Sun, Zhewei Huang, Zhichao Chang, Zhisheng Guan, Zidong Yang, Zili Zhang, Binxing Jiao, Daxin Jiang, Heung-Yeung Shum, Jiansheng Chen, Jing Li, Shuchang Zhou, Xiangyu Zhang, Xinhao Zhang, Yibo Zhu

Based on our new StepEval-Audio-360 evaluation benchmark, Step-Audio achieves state-of-the-art performance in human evaluations, especially in terms of instruction following.

Instruction Following Voice Cloning

Optimizing RLHF Training for Large Language Models with Stage Fusion

no code implementations • 20 Sep 2024 • Yinmin Zhong, Zili Zhang, Bingyang Wu, Shengyu Liu, Yukun Chen, Changyi Wan, Hanpeng Hu, Lei Xia, Ranchen Ming, Yibo Zhu, Xin Jin

Due to the intrinsic nature of RLHF training, i. e., the data skewness in the generation stage and the pipeline bubbles in the training stage, existing RLHF systems suffer from low GPU utilization.

FLUX: Fast Software-based Communication Overlap On GPUs Through Kernel Fusion

1 code implementation • 11 Jun 2024 • Li-Wen Chang, Wenlei Bao, Qi Hou, Chengquan Jiang, Ningxin Zheng, Yinmin Zhong, Xuanrun Zhang, Zuquan Song, Chengji Yao, Ziheng Jiang, Haibin Lin, Xin Jin, Xin Liu

Overall, it can achieve up to 1. 24x speedups for training over Megatron-LM on a cluster of 128 GPUs with various GPU generations and interconnects, and up to 1. 66x and 1. 30x speedups for prefill and decoding inference over vLLM on a cluster with 8 GPUs with various GPU generations and interconnects.

LoongServe: Efficiently Serving Long-Context Large Language Models with Elastic Sequence Parallelism

1 code implementation • 15 Apr 2024 • Bingyang Wu, Shengyu Liu, Yinmin Zhong, Peng Sun, Xuanzhe Liu, Xin Jin

The context window of large language models (LLMs) is rapidly increasing, leading to a huge variance in resource usage between different requests as well as between different phases of the same request.

MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs

1 code implementation • 23 Feb 2024 • Ziheng Jiang, Haibin Lin, Yinmin Zhong, Qi Huang, Yangrui Chen, Zhi Zhang, Yanghua Peng, Xiang Li, Cong Xie, Shibiao Nong, Yulu Jia, Sun He, Hongmin Chen, Zhihao Bai, Qi Hou, Shipeng Yan, Ding Zhou, Yiyao Sheng, Zhuo Jiang, Haohan Xu, Haoran Wei, Zhang Zhang, Pengfei Nie, Leqi Zou, Sida Zhao, Liang Xiang, Zherui Liu, Zhe Li, Xiaoying Jia, Jianxi Ye, Xin Jin, Xin Liu

Training LLMs at this scale brings unprecedented challenges to training efficiency and stability.

Language Modeling Language Modelling +1

Fast Distributed Inference Serving for Large Language Models

no code implementations • 10 May 2023 • Bingyang Wu, Yinmin Zhong, Zili Zhang, Shengyu Liu, Fangyue Liu, Yuanhang Sun, Gang Huang, Xuanzhe Liu, Xin Jin

Large language models (LLMs) power a new generation of interactive AI applications exemplified by ChatGPT.

Blocking Management +1

AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving

2 code implementations • 22 Feb 2023 • Zhuohan Li, Lianmin Zheng, Yinmin Zhong, Vincent Liu, Ying Sheng, Xin Jin, Yanping Huang, Zhifeng Chen, Hao Zhang, Joseph E. Gonzalez, Ion Stoica

Model parallelism is conventionally viewed as a method to scale a single large deep learning model beyond the memory limits of a single device.

Deep Learning