新闻
DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency
Hugging Face · 发布于 · 阅读约3分钟
30秒读懂
- 发生了什么
- RadixArk released DaisyRose-27B, a 27-billion-parameter model fine-tuned via reinforcement learning on long-horizon agent tasks, improving Terminal-Bench 4.1 from 5.6% to 12.0% pass rate.
- 为何重要
- Engineers building or deploying agentic systems should care, especially those working with reasoning tasks, software engineering benchmarks, or long-context agent training on limited compute budgets.
- 注意
- The model was trained on only 1,200 curated tasks over 18 days; generalization beyond these specific domains and real-world agent performance remain unvalidated by independent evaluation.
- post-train
这条新闻背后的模式
每个模式都讲清楚技术如何运作、何时值得投入,以及在哪里会失效。
The Agent Architect
每周一个模式、一个权衡、一个生产事故案例。为构建智能体系统的人准备的每周简报。
每周一封邮件,一键退订。您的地址仅用于发送简报。