Image de couverture de livre

Tongyi DeepResearch Technical Report

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Tongyi DeepResearch Team, Li, Baixuan, Zhang, Bo, Zhang, Dingchu, Huang, Fei, Li, Guangyu, Chen, Guoxin, Yin, Huifeng, Wu, Jialong, Zhou, Jingren, Li, Kuan, Su, Liangcai, Ou, Litu, Zhang, Liwen, Xie, Pengjun, Ye, Rui, Yin, Wenbiao, Yu, Xinmiao, Wang, Xinyu, Wu, Xixi, Chen, Xuanzhong, Zhao, Yida, Zhang, Zhen, Tao, Zhengwei, Zhang, Zhongwang, Qiao, Zile, Wang, Chenxi, Yu, Donglei, Fu, Gang, Shen, Haiyang, Yang, Jiayin, Lin, Jun, Zhang, Junkai, Zeng, Kui, Yang, Li, Yin, Hailong, Song, Maojia, Yan, Ming, Liao, Minpeng, Xia, Peng, Xiao, Qian, Min, Rui, Ding, Ruixue, Fang, Runnan, Chen, Shaowei, Huang, Shen, Wang, Shihang, Cai, Shihao, Shen, Weizhou, Wang, Xiaobin, Guan, Xin, Geng, Xinyu, Shi, Yingcheng, Wu, Yuning, Chen, Zhuo, Li, Zijian, Jiang, Yong
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language Artificial Intelligence Information Retrieval Machine Learning Multiagent Systems
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

_version_	1866911691349950464
author	Tongyi DeepResearch Team Li, Baixuan Zhang, Bo Zhang, Dingchu Huang, Fei Li, Guangyu Chen, Guoxin Yin, Huifeng Wu, Jialong Zhou, Jingren Li, Kuan Su, Liangcai Ou, Litu Zhang, Liwen Xie, Pengjun Ye, Rui Yin, Wenbiao Yu, Xinmiao Wang, Xinyu Wu, Xixi Chen, Xuanzhong Zhao, Yida Zhang, Zhen Tao, Zhengwei Zhang, Zhongwang Qiao, Zile Wang, Chenxi Yu, Donglei Fu, Gang Shen, Haiyang Yang, Jiayin Lin, Jun Zhang, Junkai Zeng, Kui Yang, Li Yin, Hailong Song, Maojia Yan, Ming Liao, Minpeng Xia, Peng Xiao, Qian Min, Rui Ding, Ruixue Fang, Runnan Chen, Shaowei Huang, Shen Wang, Shihang Cai, Shihao Shen, Weizhou Wang, Xiaobin Guan, Xin Geng, Xinyu Shi, Yingcheng Wu, Yuning Chen, Zhuo Li, Zijian Jiang, Yong
author_facet	Tongyi DeepResearch Team Li, Baixuan Zhang, Bo Zhang, Dingchu Huang, Fei Li, Guangyu Chen, Guoxin Yin, Huifeng Wu, Jialong Zhou, Jingren Li, Kuan Su, Liangcai Ou, Litu Zhang, Liwen Xie, Pengjun Ye, Rui Yin, Wenbiao Yu, Xinmiao Wang, Xinyu Wu, Xixi Chen, Xuanzhong Zhao, Yida Zhang, Zhen Tao, Zhengwei Zhang, Zhongwang Qiao, Zile Wang, Chenxi Yu, Donglei Fu, Gang Shen, Haiyang Yang, Jiayin Lin, Jun Zhang, Junkai Zeng, Kui Yang, Li Yin, Hailong Song, Maojia Yan, Ming Liao, Minpeng Xia, Peng Xiao, Qian Min, Rui Ding, Ruixue Fang, Runnan Chen, Shaowei Huang, Shen Wang, Shihang Cai, Shihao Shen, Weizhou Wang, Xiaobin Guan, Xin Geng, Xinyu Shi, Yingcheng Wu, Yuning Chen, Zhuo Li, Zijian Jiang, Yong
contents	We present Tongyi DeepResearch, an agentic large language model, which is specifically designed for long-horizon, deep information-seeking research tasks. To incentivize autonomous deep research agency, Tongyi DeepResearch is developed through an end-to-end training framework that combines agentic mid-training and agentic post-training, enabling scalable reasoning and information seeking across complex tasks. We design a highly scalable data synthesis pipeline that is fully automatic, without relying on costly human annotation, and empowers all training stages. By constructing customized environments for each stage, our system enables stable and consistent interactions throughout. Tongyi DeepResearch, featuring 30.5 billion total parameters, with only 3.3 billion activated per token, achieves state-of-the-art performance across a range of agentic deep research benchmarks, including Humanity's Last Exam, BrowseComp, BrowseComp-ZH, WebWalkerQA, xbench-DeepSearch, FRAMES and xbench-DeepSearch-2510. We open-source the model, framework, and complete solutions to empower the community.
format	Preprint
id	arxiv_https___arxiv_org_abs_2510_24701
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	Tongyi DeepResearch Technical Report Tongyi DeepResearch Team Li, Baixuan Zhang, Bo Zhang, Dingchu Huang, Fei Li, Guangyu Chen, Guoxin Yin, Huifeng Wu, Jialong Zhou, Jingren Li, Kuan Su, Liangcai Ou, Litu Zhang, Liwen Xie, Pengjun Ye, Rui Yin, Wenbiao Yu, Xinmiao Wang, Xinyu Wu, Xixi Chen, Xuanzhong Zhao, Yida Zhang, Zhen Tao, Zhengwei Zhang, Zhongwang Qiao, Zile Wang, Chenxi Yu, Donglei Fu, Gang Shen, Haiyang Yang, Jiayin Lin, Jun Zhang, Junkai Zeng, Kui Yang, Li Yin, Hailong Song, Maojia Yan, Ming Liao, Minpeng Xia, Peng Xiao, Qian Min, Rui Ding, Ruixue Fang, Runnan Chen, Shaowei Huang, Shen Wang, Shihang Cai, Shihao Shen, Weizhou Wang, Xiaobin Guan, Xin Geng, Xinyu Shi, Yingcheng Wu, Yuning Chen, Zhuo Li, Zijian Jiang, Yong Computation and Language Artificial Intelligence Information Retrieval Machine Learning Multiagent Systems We present Tongyi DeepResearch, an agentic large language model, which is specifically designed for long-horizon, deep information-seeking research tasks. To incentivize autonomous deep research agency, Tongyi DeepResearch is developed through an end-to-end training framework that combines agentic mid-training and agentic post-training, enabling scalable reasoning and information seeking across complex tasks. We design a highly scalable data synthesis pipeline that is fully automatic, without relying on costly human annotation, and empowers all training stages. By constructing customized environments for each stage, our system enables stable and consistent interactions throughout. Tongyi DeepResearch, featuring 30.5 billion total parameters, with only 3.3 billion activated per token, achieves state-of-the-art performance across a range of agentic deep research benchmarks, including Humanity's Last Exam, BrowseComp, BrowseComp-ZH, WebWalkerQA, xbench-DeepSearch, FRAMES and xbench-DeepSearch-2510. We open-source the model, framework, and complete solutions to empower the community.
title	Tongyi DeepResearch Technical Report
topic	Computation and Language Artificial Intelligence Information Retrieval Machine Learning Multiagent Systems
url	https://arxiv.org/abs/2510.24701