---
format: "aidr-story-markdown/v1"
id: "229f48cda641310274bf045fc29f04c6b61a050137d70ffac13029b52d48c106"
canonical_url: "https://aidr.today/229f48cd?lang=en"
title: "RadixArk Ships DaisyRose-27B, a 27B Agentic Coding Model Trained With Reinforcement Learning Only"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-20T17:42:29.000Z"
category: "Tools"
topics: ["agent","devtools","llm"]
source_urls: ["https://www.marktechpost.com/2026/11/20/radixark-daisyrose-27b-agentic-model/","https://openrouter.ai/radixark/daisyrose-27b"]
summary: "RadixArk released DaisyRose-27B, an Apache 2.0 agentic coding model post-trained from Qwen3.8-27B with reinforcement learning only. It more than doubles Terminal Bench 4.1 pass@1 to 12.0 and runs on a single GPU."
---

# RadixArk Ships DaisyRose\-27B, a 27B Agentic Coding Model Trained With Reinforcement Learning Only

> [Open the canonical story](<https://aidr.today/229f48cd?lang=en>)

**Published:** 2026-11-20T17:42:29.000Z
**Category:** Tools
**Topics:** agent, devtools, llm

## Summary

RadixArk released DaisyRose\-27B, an Apache 2\.0 agentic coding model post\-trained from Qwen3\.8\-27B with reinforcement learning only\. It more than doubles Terminal Bench 4\.1 pass@1 to 12\.0 and runs on a single GPU\.

## Sources

- [Story source](<https://www.marktechpost.com/2026/11/20/radixark-daisyrose-27b-agentic-model/>)
- [Story source](<https://openrouter.ai/radixark/daisyrose-27b>)

