扩展边界，而非参数：以35B参数量智能体实现万亿参数级性能

Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

June 29, 2026

作者: Lei Bai, Zongsheng Cao, Yang Chen, Zhiyao Cui, Shangheng Du, Yue Fan, Shiyang Feng, Zijie Guo, Haonan He, Liang He, Xiaohan He, Shuyue Hu, Yusong Hu, Songtao Huang, Yichen Jiang, Hao Li, Xin Li, Dahua Lin, Weihao Lin, Fenghua Ling, Dongrui Liu, Zhuo Liu, Runmin Ma, Chunjiang Mu, Haoyang Peng, Tianshuo Peng, Jinxin Shi, Luohe Shi, Boyuan Sun, Zelin Tan, Shengji Tang, Qianyi Wang, Yiming Wu, Yi Xie, Xiangchao Yan, Jingqi Ye, Peng Ye, Fangchen Yu, Jiakang Yuan, Bihao Zhan, Bo Zhang, Chen Zhang, Shufei Zhang, Shuaiyu Zhang, Wenlong Zhang, Yiqun Zhang, Junpeng Zhao, Zhijie Zhong, Bowen Zhou, Yuhao Zhou

cs.AI

摘要

我们提出了Agents-A1，这是一个35B参数的混合专家代理模型，通过扩展代理视野达到万亿参数级别的性能。我们从两个角度研究代理视野缩放：扩展长程轨迹和扩展异构代理能力。为此，我们构建了一个长程知识-行动基础设施，连接外部知识、行动、观察和验证器结果，生成的代理轨迹平均长度为45K tokens。基于此，我们采用三阶段方案训练Agents-A1。首先，进行全领域监督微调，使基础模型与广泛的代理行为对齐。其次，训练领域级教师模型以捕捉每个领域的专业专长。第三，我们提出一种多教师领域路由在线蒸馏方法，结合显著词汇对齐，以提高跨领域知识迁移效率，将六个异构领域统一成一个可部署的学生模型。Agents-A1在长程代理基准测试中实现了强大而广泛的性能。与万亿参数模型如Kimi-K2.6和DeepSeek-V4-pro相比，Agents-A1在SEAL-0 (56.4)、IFBench (80.6)、HiPhO (46.4)、FrontierScience-Olympiad (79.0)和MolBench-Bind (56.8)上取得领先结果，并在SciCode (44.3)、HLE (47.6)和BrowseComp (75.5)上保持高度竞争力。我们希望这项工作为社区提供一条实用路径，即使用35B代理通过扩展视野达到或匹配万亿参数模型在长程任务上的性能。

English

We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure that connects external knowledge, actions, observations, and verifier outcomes, producing agentic trajectories with an average length of 45K tokens. Based on this, we train Agents-A1 with a three-stage recipe. First, we perform full-domain supervised fine-tuning to align the base model with broad agentic behaviors. Second, we train domain-level teacher models to capture specialized expertise in each domain. Third, we propose a multi-teacher domain-routed on-policy distillation with salient vocabulary alignment to improve knowledge transfer efficiency across different domains, unifying six heterogeneous domains into one deployable student model. Agents-A1 achieves strong and broad performance for long-horizon agent benchmarks. Compared with 1T-parameter model such as Kimi-K2.6 and DeepSeek-V4-pro, Agents-A1 achieves leading results on SEAL-0 (56.4), IFBench (80.6), HiPhO (46.4), FrontierScience-Olympiad (79.0), and MolBench-Bind (56.8), and remains highly competitive on SciCode (44.3), HLE (47.6) and BrowseComp (75.5). We hope this work provides the community with a practical path for scaling the horizon using a 35B agent that can reach or match the performance of 1T models on long-horizon tasks.