agentic-modeling

Microsoft's Orchard is an open-source framework for agentic modeling, providing a Kubernetes-native environment substrate, RL training stack, and datasets for software engineering, GUI, and computer use agents.

The tweet highlights the potential of agent harnesses to capture and distill experts' tacit knowledge as new training data, referencing a Berkeley AI Summit talk by Jianfeng Gao on agentic modeling as an emerging AI paradigm.

作者在Berkeley AI Summit 2026总结了Jianfeng Gao教授的演讲，指出AI建模的新范式是基于Agentic Modeling和新数据飞轮，强调Harness将成为应用护城河的关键。

微软研究院与合作伙伴推出 Orchard，一个用于可扩展代理建模的开源框架，具有轻量级的 Kubernetes 原生沙箱环境。它使用小模型在 SWE-bench Verified（67.5%）和 GUI 基准测试（平均 68.4%）上取得了最先进的结果。

Orchard是一个用于可扩展智能体建模的开源框架，能够训练多样化的自主智能体，在编程、GUI导航和个人辅助任务上取得了最先进的结果。