DeepInfra raises $107M Series B to scale the inference cloud — read the announcement
RadixArk/
$0.15
in
$1.875
out
$0.038
cached
/ 1M tokens
$0.20
in
$2.50
out
$0.05
cached
/ 1M tokens
DaisyRose-27B is RadixArk's multimodal reasoning model, trained from Qwen3.8-27B with reinforcement learning only on long-horizon agent tasks. It improves on the base model in agentic coding, reasoning and instruction following, with thinking that can be enabled or disabled, image understanding and a 256K-token context.

Ask me anything
Settings
DaisyRose-27B is RadixArk's first model: a multimodal reasoning model built on Qwen3.8-27B with reinforcement learning only, without any supervised fine-tuning or distillation. It was trained on about 1,200 curated long-horizon agent tasks, with sessions of up to 1M tokens, for 18 days on 144 H200 GPUs. Terminal Bench 4.1 pass@1 rises from 5.6 to 12.0, and the model also improves on most of the other coding, reasoning and instruction-following benchmarks below.
DaisyRose-27B keeps the base model's architecture, tokenizer and chat template, along with its image understanding and 256K-token context, so it is a drop-in replacement for Qwen3.8-27B.
reasoning_effort sets its depth.| Property | Value |
|---|---|
| Model type | Multimodal causal language model with a vision encoder; thinking model |
| Architecture | Qwen3.8-27B: 64 decoder layers in a 3:1 pattern of linear-attention (Gated DeltaNet) and full-attention layers, hidden size 5,120, a 27-layer vision encoder, one multi-token-prediction layer |
| Parameters | 27.8B in total, including a 0.46B vision encoder and a 0.42B multi-token-prediction head |
| Context length | 262,144 tokens |
| Input / output | Text and images / text |
| Served on DeepInfra | FP8 weights with DSpark speculative decoding, on SGLang |
DaisyRose-27B against its base model, open models of similar or larger size, and GPT-6 Luna. Scores as reported by RadixArk; the best score in each row is highlighted.
| DaisyRose-27B | Qwen3.8-27B | Muse Glimmer-30B | Qwen3.6-35B-A3B | MiMo V2.6 Flash | Inkling 2 Small | Beam 1.0 | Muse Spark Open | GPT-6 Luna | |
|---|---|---|---|---|---|---|---|---|---|
| Size | |||||||||
| Activated parameters | 27B | 27B | 30B | 3B | 15B | 17B | 23B | 42B | – |
| Total parameters | 27B | 27B | 30B | 35B | 309B | 594B | 501B | 816B | – |
| Agentic coding | |||||||||
| Terminal Bench 4.1 | 12.0 | 5.6 | 0.5 | 0.0 | 16.1 | 13.0 | 6.4 | 6.6 | 15.2 |
| Terminal Bench 2.1 | 75.5 | 73.0 | 51.7 | 44.9 | 87.6 | 81.3 | 80.1 | 76.4 | 73.0 |
| DeepSWE v1.1 | 46.2 | 42.2 | 4.4 | 0.0 | 67.9 | 72.1 | 43.3 | 60.0 | 66.6 |
| SWE Bench Pro v2 | 94.7 | 91.3 | 80.4 | 73.3 | – | 97.7 | – | 95.0 | – |
| Reasoning | |||||||||
| HLE (text only) | 34.3 | 33.9 | 22.0 | 22.2 | 35.1 | 44.1 | 34.7 | 43.9 | 38.5 |
| GPQA Diamond | 90.7 | 89.2 | 83.5 | 86.0 | 88.7 | 93.0 | 90.6 | 89.0 | 90.5 |
| Instruction following | |||||||||
| IFBench | 82.4 | 79.5 | 77.0 | 64.4 | – | – | 79.6 | – | – |
All DaisyRose-27B runs use thinking mode at temperature 1.0, top_p 0.95, top_k 20, min_p 0, reasoning effort xhigh, with no max_tokens cap within the 256K context. Terminal Bench 4.1: 8 repetitions per task, mini-swe-agent harness with summarization. Terminal Bench 2.1: 5 repetitions, Terminus 2 harness. DeepSWE v1.1: 8 repetitions, mini-swe-agent. SWE Bench Pro v2: 5 repetitions, mini-swe-agent, full 642-task set. HLE: 5 repetitions, no tools. IFBench: 4 repetitions, prompt-level loose accuracy. GPQA Diamond: 8 repetitions, answer order reshuffled each time. Other models' numbers are as reported by their developers or leaderboards; see the original model card for every source.
temperature 1.0, top_p 0.95, top_k 20, min_p 0.content and the thinking in reasoning_content. In multi-turn chats, send earlier turns back with their reasoning_content.reasoning_effort: low, medium, or high (the model's deepest level, also the default). none turns thinking off.max_tokens unset for hard problems: long thinking can run to tens of thousands of tokens.DaisyRose-27B is trained from Qwen3.8-27B with reinforcement learning only; the vision encoder and the multi-token-prediction head are the base model's, unchanged. The tasks were built with Eigent AI and each one reviewed together with the model's attempts. Training ran fully asynchronously with Miles, RadixArk's open-source reinforcement-learning framework: 128 H200s ran the agent on its tasks, served by SGLang, while 16 more trained on the finished trajectories, about 63,000 GPU-hours in all.
DaisyRose-27B is released under the Apache 2.0 license. It is a derivative of Alibaba's Qwen3.8-27B, which is also licensed under Apache 2.0.
© 2026 DeepInfra. All rights reserved.