news
26

RadixArk has introduced DaisyRose-27B, its first model, trained with Miles on 1,200 curated tasks over 18 days using 144 H200 GPUs. The reinforcement learning run improved Qwen3.8 from 5.6% to 12% on Terminal-Bench 4.1.

Only full members can comment

Log in to express your opinion.

No comments yet.