Skip to content
Thursday, 19 November 2026
Tenesys AI News
Subscribe

DaisyRose-27B

Change history

  1. 18 Nov 2026
    RadixArk Releases DaisyRose-27B Agent-Focused Model
    ParameterBeforeNow
    Terminal-Bench 4.1 pass@15.6%12.0%
    Context window256K256K

News

Models· Important· 🧪 Worth Testing

RadixArk Releases DaisyRose-27B Agent-Focused Model

RadixArk has launched DaisyRose-27B, a 27-billion parameter model built on Qwen3.8-27B, specifically fine-tuned for complex long-horizon agent tasks using reinforcement learning. The model demonstrates significant improvements across various reasoning and coding benchmarks, particularly in terminal-based automation.

Hugging Face
DaisyRose-27B · TENESYS AI NEWS