[论文] Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work

论文概要

研究领域: ML 作者: Wenhui Chen, Shiwen Cheng, Hao Dong, Chenda Duan, Ruixiang Feng, Zhong Guan, Boqiang Guo, Xueyuan Han, Haojie Hao, Liangmeng Huang, Zhelong Huang, Xinke Kong, Hongyu Li, Jiazheng Li, Junbo Li, Qingchuan Li, Yukun Lian, Chang Liu, Tianyu Liu, Zicheng Liu, Shuyi Ouyang, Yijun Pan, Kunyu Shi, Xiaojun Tang, Bingquan Wang, Kesu Wang, Yuchen Wang, Sibo Wei, Sicong Xie, Xiaoying Xing, Yi Xu, Zhijun Xu, Hongwei Xue, Qingcheng Zeng, Di Zhang, Guannan Zhang, Haochen Zhang, Tianlong Zhang, Tianyu Zhao, Tianyu Zhao, Yanjun Zheng, Jialong Zhu, Zijian Zou 发布时间: 2026-09-15 arXiv: 2609.11977

中文摘要

协同工作智能体执行融合信息收集、工具使用、代码编写与文件操作的复杂工作流，跨越多次模型调用。由于成本与延迟在整个任务周期中累积，其实际价值不仅取决于峰值能力，更取决于能力交付的效率。而日常工作中的许多步骤强调的是状态追踪、协调、恢复与跟进，而非前沿规模的推理。我们提出 Occamy-1.0——一个成本高效的协同工作模型，通过对后训练完成的 Qwen3.6-35B-A3B 检查点进一步训练获得。我们构建基于执行的数据与环境，跨多种 harness 捕获可回放的长程轨迹，并使用分阶段后训练来发展并巩固互补的执行能力。在广泛的协同工作基准上，Occamy-1.0 始终位居同规模最强模型之列，并在若干任务上与规模大得多的前沿系统保持竞争力。在我们规定的评估与定价协议下，其四个代表性基准的综合表现使其处于观测到的成本-性能帕累托前沿的低成本拐点。工具调用、编程与指令遵循方面的支持性评估进一步表明，这种专业化保留了广泛的智能体能力。我们开源模型权重与部分训练数据，以支持对实用协同工作智能体与智能体后训练的研究。

原文摘要

Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations. Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability is delivered. Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning. We present Occamy-1.0, a cost-efficient co-work model obtained by further training the post-trained Qwen3.6-35B-A3B checkpoint. We construct execution-grounded data and environments, capture replayable long-horizon trajectories across multiple harnesses, and use staged post-training to develop and consolidate complementary execution cap...

#论文 #arXiv #ML #小凯