Orchard：一个开源的智能体建模框架
Orchard: An Open-Source Agentic Modeling Framework
AI总结：
本文提出Orchard，一个开源的智能体建模框架，通过轻量级环境服务和三种智能体建模食谱，实现了跨领域可重用的智能体数据、训练和评估。
摘要：
智能体建模旨在通过规划、推理、工具使用和与环境的多轮交互，将大语言模型转化为能够解决复杂任务的自主智能体。尽管有大量投入，开放研究仍受制于基础设施和训练差距。许多高性能系统依赖于专有代码库、模型或服务，而大多数开源框架专注于编排和评估，而非可扩展的智能体训练。我们提出了Orchard，一个用于可扩展智能体建模的开源框架。其核心是Orchard Env，一个轻量级环境服务，提供可重用的原语用于跨任务领域、智能体利用和流水线阶段的沙盒生命周期管理。在Orchard Env之上，我们构建了三种智能体建模食谱。Orchard-SWE针对编码智能体，从MiniMax-M2.5和Qwen3.5-397B中蒸馏出107K条轨迹，引入信用分配SFT来学习未解决轨迹的 productive 段落，并应用平衡自适应回滚进行强化学习。从Qwen3-30B-A3B-Thinking开始，Orchard-SWE在SWE-bench Verified上经过SFT后达到64.3%，经过SFT+RL后达到67.5%，在同等规模的开源模型中设立了新的状态。Orchard-GUI使用仅0.4K蒸馏轨迹和2.2K开放性任务训练了一个4B视觉-语言计算机使用智能体，在WebVoyager、Online-Mind2Web和DeepShop上分别达到74.1%、67.0%和64.0%的成功率，成为最强的开源模型，同时在与专有系统竞争中保持竞争力。Orchard-Claw针对个人助理智能体，仅用0.2K合成任务训练，达到Claw-Eval上的59.6% pass@3和与更强的ZeroClaw利用配合时的73.9%。总体而言，这些结果表明，一个轻量级、开源、不依赖利用的环境层能够实现跨领域的可重用智能体数据、训练和评估。
英文摘要：
Agentic modeling aims to transform LLMs into autonomous agents capable of solving complex tasks through planning, reasoning, tool use, and multi-turn interaction with external environments. We present Orchard, an open-source framework for scalable agentic modeling. At its core is Orchard Env, a lightweight Kubernetes-native environment service that provides reusable primitives for sandbox lifecycle management across task domains, agent harnesses, and training stages. On top of Orchard Env, we build three agentic modeling recipes. Orchard-SWE targets software engineering agents. We introduce credit-assignment supervised fine-tuning and a progression of RL signals: Balanced Adaptive Rollout (BAR) for sparse-reward optimization, on-policy distillation (OPD) and rubric-based process reward (RPR) for dense supervision, and historical experience distillation, which compresses rollouts from prior experiments into a compact value model for inference-time reranking. Built on the Qwen3.5-35B-A3B backbone, Orchard-SWE reaches 69.7% with RPR-based RL and 73.0% with value-model reranking on SWE-bench Verified, setting a new state of the art among open-source methods while approaching frontier systems over 10x larger. Orchard-GUI trains a 4B vision-language computer-use agent using only 0.4K distilled trajectories and 2.2K open-ended tasks, achieving 68.4% average success across WebVoyager, Online-Mind2Web, and DeepShop, making it the strongest open-source model while remaining competitive with proprietary systems. Orchard-Claw targets personal assistant agents. Trained with only 0.2K synthetic tasks, it achieves 59.6% pass@3 on Claw-Eval and 73.9% when paired with the stronger ZeroClaw harness. Collectively, these results demonstrate that a lightweight, open, harness-agnostic environment layer enables reusable agentic data, training recipes, and evaluation protocols across domains.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
