沙丘社区
论文

LongCat-Flash-Thinking-2601技术报告

LongCat-Flash-Thinking-2601 Technical Report

Meituan LongCat Team, Anchun Gui, Bei Li, Bingyang Tao, Bole Zhou, Borun Chen, Chao Zhang, Chao Zhang, Chen Gao, Chen Zhang, Chengcheng Han, Chenhui Yang, Chuyu Zhang, Cong Chen, Cunguang Wang, Daoru Pan, Defei Bu, Dengchang Zhao, Di Xiu, Dishan Liu, Dongyu Ru, Dunwei Tu, Fan Wu, Fengcheng Yuan, Fengcun Li, Gang Xu, Guanyu Wu, Guoyuan Lin, Haibin Wang, Hansi Yang, Hao Yang, Haonan Yan, Haoxiang Ma, Haoxing Wen, Hongyan Hao, Hongyin Tang, Hongyu Zang, Hongzhi Ni, Hui Su, Jiacheng Zhang, Jiahong Zhou, Jiahuan Li, Jiaming Wang, Jian Yang, Jianfei Zhang, Jianhao Xu, Jianing Wang, Jiapeng Zhu, Jiaqi Sun, Jiarong Shi, Jiarui Zhao, Jingang Wang, Jinluan Yang, Jinrui Ding, Jinwei Xiao, Jiyuan He, Juncan Xu, Kefeng Zhang, Keheng Wang, Li Wei, Lianhui Ma, Lin Qiu, Lingbing Kong, Lingchuan Liu, Linsen Guo, Mengshen Zhu, Mengxia Shen, Mingyang Zhu, Peiguang Li, Peng Pei, Pengcheng Jia, Pengtao Zhang, Peng Zhao, Qi Gu, Qiong Huang, Qiyuan Duan, Quanchi Weng, Rongxiang Weng, Rongzhi Zhang, Rumei Li, Shanglin Lei, Shengnan An, Shijun Dai, Shuaikang Liu, Shuang Zhou, Shuo Wang, Songyuan Zhao, Tao Liang, Tianhao Hu, Tianze Chen, Wei Liu, Wei Shi, Wei Wang, Weifeng Tang, Wenjie Shi, Wenlong Zhu, Wentao Chen, Wentao Shi, Xi Su, Xiangcheng Liu, Xiandi Ma, Xiangyu Xi, Xiangyuan Liu, Xiangzhou Huang, Xiao Liu, Xiaodong Cai, Xiaolong Chen, Xiaowei Shi, Xiaoyu Li, Xin Chen, Xingchen Liu, Xuan Huang, Xuezhi Cao, Xunliang Cai, Yan Chen, Yang Bai, Yang Liu, Yang Yang, Yang Zheng, Yaoming Wang, Yaoming Zhu, Yaqi Huo, Yanyu Chen, Yaorui Shi, Yerui Sun, Yi Zhang, Yihao Chen, Yi-Kai Zhang, Yifan Lu, Yifan Zhao, Yitao Zhai, Yongjing Yin, Yongwei Zhou, Youshao Xiao, Yuchuan Dai, Yuchen Xie, Yuchen Yu, Yufei Zhang, Yuhuai Wei, Yulei Qian, Yunfan Liang, Yunke Zhao, Yuwei Jiang, Yuxin Bian, Yuxin Chen, Yuxin Liu, Yue Xu, Yueqing Sun, Zeyang Yu, Zhao Yang, Zhengsheng Huang, Zhengyu Chen, Zhijian Liu, Zhikang Xia, Zhimin Lin, Zhiyuan Yao, Zhuofan Chen, Zhuowen Han, Zijian Zhang, Ziran Li, Ziwen Wang, Ziyuan Zhuang
模型训练模型架构MoE强化学习Agentic RL

摘要

我们推出 LongCat-Flash-Thinking-2601，一个 5600 亿参数的开源混合专家（MoE）推理模型，具备卓越的 agentic 推理能力。LongCat-Flash-Thinking-2601 在广泛的 agentic 基准（包括 agentic 搜索、agentic 工具使用与工具集成推理）上于开源模型中取得最先进性能。除基准表现外，该模型展现出对复杂工具交互的强泛化能力以及在嘈杂真实世界环境中的稳健行为。其先进能力源于一个统一训练框架：把领域并行专家训练与后续融合相结合，并覆盖从预训练到后训练的数据构建、环境、算法与基础设施的端到端协同设计。特别是，模型在复杂工具使用上的强泛化能力来自我们对环境扩展与有原则的任务构建的深入探索。为优化长尾、偏斜生成与多轮 agentic 交互，并在跨越 20 多个领域的 10,000 多个环境上实现稳定训练，我们系统扩展了异步强化学习框架 DORA，以实现稳定高效的大规模多环境训练。此外，认识到真实任务本质上有噪声，我们对真实世界噪声模式进行系统分析与分解，并设计有针对性的训练流程，把此类不完美显式纳入训练过程，从而提升真实应用的鲁棒性。为进一步提升复杂推理任务性能，我们引入 Heavy Thinking 模式，通过密集并行思考同时扩展推理深度与宽度，实现有效的测试时扩展。

LongCat-Flash-Thinking-2601技术报告
图1：LongCat-Flash-Thinking-2601 的基准测试性能。