Subscribe

DaisyRose-27B

RadixArk released DaisyRose-27B, its first model, a Qwen3.8-27B fine-tune trained with reinforcement learning that raises Terminal-Bench 4.1 pass@1 from 5.6% to 12.0%.

RadixArk trained the model with its open-source Miles framework on about 1,200 curated long-horizon agent tasks, with sessions up to 1M tokens, in 18 days on 144 H200 GPUs. It says its internal training backend reached 37% MFU against 16% for the open-source trainer, which cuts GPU time by 57%. BF16, FP8 and NVFP4 checkpoints are available on Hugging Face.

Date
Wednesday 18 November 2026
Lab
RadixArk
Kind
open-weights
Access
open weights
Price
$0.15 per M input, $1.875 per M output on OpenRouter (DeepInfra, listed 25% off $0.20 / $2.50)

Figures

MeasureValueMeasured by
Terminal-Bench 4.1 pass@112.0% (base 5.6%)
averaged over 8 runs per task, internal mini-swe-agent harness with summarization
company
Training MFU37% vs 16% for open-source trainer
57% less GPU time for the same training
company
Training compute63,000 GPU-hours on 144 H200scompany
Sandbox costabout 80% lower than popular cloud sandbox providers
about 944,000 sandboxes run
company

Built on Qwen3.8-27B with training tasks made with Eigent AI. The license was not stated on the page.

Sources

  1. radixark.com/blog/daisyrose-27b
  2. openrouter.ai/radixark/daisyrose-27b

This record was checked against its sources on 19 November 2026. How we check

Read the daily brief for 18 November 2026