Skip to content
Tagged articles

Miles

1 articles · Page 1 of 1
SmartApe AI
SmartApe AI
Nov 20, 2026 · Artificial Intelligence

Ex-xAI & NVIDIA Engineers Build DaisyRose-27B on 144 H200s with Custom RL Infra

RadixArk, founded by ex-xAI inference lead Sheng Ying and ex-NVIDIA researcher Zhu Banghua, releases DaisyRose-27B, a 27B parameter model trained on 144 H200 GPUs in 18 days using their custom RL framework Miles and SGLang inference engine, achieving 2.14x pass@1 improvement on Terminal-Bench 4.1.

AI infrastructureDaisyRose-27BH200
0 likes · 9 min read
Ex-xAI & NVIDIA Engineers Build DaisyRose-27B on 144 H200s with Custom RL Infra