DaisyRose-27B: RL on 144 H200s for 18 days lifts Terminal-Bench from 5.6% to 12.0%

ying11231 · x · 2026-11-19

RadixArk released its first open-weight model, DaisyRose-27B: RL trained on Qwen3.8-27B with 1,200 curated long-horizon agent tasks across 18 days on 144 H200s, stretching agent sessions to 1M tokens.

Key results:

A concrete proof that small models + curated RL tasks can close much of the gap to larger models, with the training recipe and engineering data shared publicly.

Related event: RadixArk open-sources DaisyRose-27B: pure RL training doubles Terminal-Bench score(7 posts)→

Original post →

More from Models

Models channel →