← Back to the wire

DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency

AnnouncementModelNov 18, 2026

RadixArk's DaisyRose-27B improves Terminal-Bench 4.1 pass@1 from 5.6% to 12.0% over its base model Qwen3.8-27B. Trained with reinforcement learning on 1,200 curated long-horizon agent tasks over 18 days on 144 H200 GPUs, the model also gains on other reasoning, instruction-following, and coding benchmarks. Checkpoints are available under Apache 2.0. DaisyRose-27B retains Qwen3.8-27B's architecture, tokenizer, and 256K-token context window.

Receipt № 29571 source · awaiting confirmation ◐

Evidence

1source· awaiting independent confirmation

No score is assigned. Sources and their independence are shown in the citation chain below.

Citation chain · 1 source

01med
huggingface.coBloghttps://huggingface.co/blog/RadixArk/daisyrose-27b-pushing-the-frontier-of-post-traininThis is RadixArk's own release announcement, giving firsthand detail on the training pipeline, the Miles framework's efficiency claims, and the sandbox infrastructure behind DaisyRose-27B.
PRIMARY
OpenAICompanyRadixArkCompanyQwen3.8-27BModelDaisyRose-27BModel
Canonical: https://huggingface.co/blog/RadixArk/daisyrose-27b-pushing-the-frontier-of-post-trainin