Skip to content
FluxCDNFluxCDN
esc

RadixArk: DaisyRose 27B

DaisyRose 27B is an open-weight multimodal reasoning model from RadixArk, trained from Qwen3.8 27B using reinforcement learning only on long-horizon agent tasks, with no supervised fine-tuning or distillation.

Input
54,656 Toman/M tokens
Output
676,368 Toman/M tokens
Context
262K
Max output
236K
Published
Nov 21, 2026
Maker
RadixArk

Free plan prices, in Toman at today's rate. Plans and prices

Price by plan

Toman per million tokens at today's rate. A plan lowers the price of your usage up to a limit each period.

PlanInputOutput
Free54,656676,368
Pro52,416648,648
Max50,176620,928

Capabilities

  • Text
  • Vision
  • Tools
  • JSON
  • Reasoning

What it takes and gives

Accepts
  • Text
  • Image
Creates
  • Text

Every input listed here works through the API and the playground: images and PDFs as data URLs or https links, audio as base64, video as a data URL or a link the model supports.

Call it

Pass radixark/daisyrose-27b as the model name, or let router/auto choose for you.

import os

from openai import OpenAI

client = OpenAI(
    base_url="https://ai.fluxcdn.cloud/api/v1",
    api_key=os.environ["FLUXCDN_API_KEY"],
)

resp = client.chat.completions.create(
    model="router/auto",
    messages=[{"role": "user", "content": "Hello"}],
)

print(resp.choices[0].message.content)