Microsoft Research· Baolin Peng, Wenlin Yao, Qianhui Wu, Hao Cheng, Jianfeng Gao·· 2026-08-04
Orchard: An open framework for scalable agentic AI
Orchard: An open framework for scalable agentic AI
AI summary
Orchard is an open-source framework for the research community to train and evaluate AI agents across task types. It reduces complexity while supporting strong performance from smaller models by enabling researchers to reuse the same infrastructure.
Selection record
Threshold 60Official, first-handFirst 71Second 71
AdmittedSum of both 142 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：开源智能体AI框架与训练评测
- Why it was chosen
- The open-source environment service, training workflow and data explain how small models can be trained inside real deployment frameworks.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Microsoft Research · microsoft.com