Zlio
Everything in AI and robotics, on one page.
Updated 01:32 UTC · Next update 02:00 UTC
Zlio Daily, the newsletter, is coming soon. The day's essential AI and robotics news in your inbox. Leave your email and we'll tell you when it launches.
One email when we launch. No spam, unsubscribe anytime. Privacy
Open Models · first seen 18 November 2026, 19:39 UTC

RadixArk releases DaisyRose-27B agent-training model

Shows post-training efficiency gains for agentic coding benchmarks on an open Qwen base.

DaisyRose-27B is a 27B-parameter model from RadixArk built to showcase its agent training infrastructure. A 1,200-task RL run reportedly lifted Qwen3.8 on Terminal-Bench 4.1 from 5.6% to 12%.

RadixArkOpenOpen weights
Name
DaisyRose-27B
Params
27B
Domain
software

Read it at Hugging Face Blog → 💬 32 on HN

Share LinkedIn X Bluesky WhatsApp

Coverage: 3 sources, in order

  1. Hacker News Read the article →
  2. Hugging Face Blog Read the article →
  3. RuntimeWire Read the article →

← Front page