titleStepAudio 3 Realtime Technical Report
authorsBin Lin, Bo Zhao, Boyang Zhang, Boyong Wu, Chao Yan, Chen Geng, Chen Wu, Cheng Yi, Chengli Feng, Chenglin Zhu, Chengting Feng, Chengyuan Yao, Daijiao Liu, DanNi Wan, Daxin Jiang, Dongjian Li, Dongqing Pang, Fei Tian, Feng Tian, Future Li, Gang Yu, Guanglong Yang, Haoyang Zhang, Hongyuan Wang, Jia Peng, Jiahao Song, Jialong Xue, Jiamin Fan, Jiangjie Zhen, Jianzheng Gao, Jincheng Wen, Jinghua Liang, Jinglan Gong, Jun Chen, Li Xie, Liang Zhao, Lifang Zhang, Lingli Ji, Lun Cai, Min Xu, Peilin Li, Peng Yang, Pengfei Tan, Qingjian Lin, Qinxin Du, Ruijie Xiong, Runze Li, Shenghua Hu, Shengqian Qin, Shi Qiu, Siqi Tu, Siyi Zhou, Tianjiao Deng, Wanying Lu, Weiming Niu, Wen Sun, WenWen Qu, Xiangyu Zhang, Xianwei Zhang, Xiaosu Su, Xing Chen, Xinyu Liu, Xuerui Yang, Yan Wu, Yang Li, Yang Yang, Yechang Huang, Yibo Zhu, Yifan Zhang, Yinuo Yan, Youjun Chen, Yu Fu, Yu Luo, Yu Zhou, Yujie Chen, Yumang Wang, Yunzhou Ju, Yuxiang Yang, Yuxin Li, Yuxin Zhang, Zekai Liu, Zengwei Yao, Zhaoxin Yuan, Zhenwei Mou, Zhiquan Zhang, Zhiyue Wu, Zichao Li, Zichao Zhou, Ziqi Ren, Zixuan Wang
date2026-09-12
venuearXiv (preprint, not peer reviewed)
idarXiv:2609.14005
affiliationStepFun
upvotes120 (Hugging Face Daily Papers, 2026-11-01)
linksabs pdf hf
abstractRealtime spoken interaction demands deep reasoning, prompt responses, and fluid turn-taking. We present StepAudio 3 Realtime, an audio-language foundation model organized around a continuous listen-converse-think-act loop. Deep Perception captures rich acoustic cues to interpret user intent, while Seamless Duplex models synchronized audio streams to handle pauses, backchannels, and interruptions naturally. Crucially, we resolve the tension between deep deliberation and latency via Think-While-Speaking, executing private reasoning in parallel with spoken delivery. In reasoning mode, StepAudio 3 reaches a 73.0 macro average on StepAudioChat. With Think-While-Speaking, it achieves dialogue and reasoning performance comparable to dedicated reasoning models while speaking in real time. Furthermore, an integrated Voice Agent handles asynchronous tool execution without disrupting the dialogue flow. StepAudio 3 Realtime achieves top-tier performance across key dimensions: an exceptional 90.6 on the MMSU benchmark, 98.9 Overall on the Artificial Analysis Full-Duplex Bench, and a 56.0% macro task-success rate on τ-Voice.

cite

@misc{lin2026stepaudio,
  title={StepAudio 3 Realtime Technical Report},
  author={Bin Lin and Bo Zhao and Boyang Zhang and Boyong Wu and Chao Yan and Chen Geng and Chen Wu and Cheng Yi and Chengli Feng and Chenglin Zhu and Chengting Feng and Chengyuan Yao and Daijiao Liu and DanNi Wan and Daxin Jiang and Dongjian Li and Dongqing Pang and Fei Tian and Feng Tian and Future Li and Gang Yu and Guanglong Yang and Haoyang Zhang and Hongyuan Wang and Jia Peng and Jiahao Song and Jialong Xue and Jiamin Fan and Jiangjie Zhen and Jianzheng Gao and Jincheng Wen and Jinghua Liang and Jinglan Gong and Jun Chen and Li Xie and Liang Zhao and Lifang Zhang and Lingli Ji and Lun Cai and Min Xu and Peilin Li and Peng Yang and Pengfei Tan and Qingjian Lin and Qinxin Du and Ruijie Xiong and Runze Li and Shenghua Hu and Shengqian Qin and Shi Qiu and Siqi Tu and Siyi Zhou and Tianjiao Deng and Wanying Lu and Weiming Niu and Wen Sun and WenWen Qu and Xiangyu Zhang and Xianwei Zhang and Xiaosu Su and Xing Chen and Xinyu Liu and Xuerui Yang and Yan Wu and Yang Li and Yang Yang and Yechang Huang and Yibo Zhu and Yifan Zhang and Yinuo Yan and Youjun Chen and Yu Fu and Yu Luo and Yu Zhou and Yujie Chen and Yumang Wang and Yunzhou Ju and Yuxiang Yang and Yuxin Li and Yuxin Zhang and Zekai Liu and Zengwei Yao and Zhaoxin Yuan and Zhenwei Mou and Zhiquan Zhang and Zhiyue Wu and Zichao Li and Zichao Zhou and Ziqi Ren and Zixuan Wang},
  year={2026},
  eprint={2609.14005},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2609.14005}
}