Qwen3.8-27B
Language modelOpen weightsCard summarizedA 27B native vision-language dense model of the Qwen3.8 series
ActiveSource: Qwen3.8-27B model card (Hugging Face)
Use it via API
$309 /month
$6.18 per 1K requests
RAG over docs: 12K in · 600 out · 50K requests/mo · 30% cached · change
Where to call it
- Alibaba Cloud Model Studio
qwen3.8-27b - Self-host (weights)
Qwen/Qwen3.8-27B
Good for
- Computer use (OSWorld-Verified 84.3%) and mobile use (AndroidWorld 81.9%)
- Agentic coding for its size (SWE-bench Pro 61.7%, Terminal Bench 2.1 73.0%)
- Document understanding (OmniDocBench 1.5 91.1%)
Overview
Alibaba · 04Open, front
Qwen3.8-27B
A 27B native vision-language dense model of the Qwen3.8 series
- Cost
- $0.50 in
- Context
- 1M
- Input / output
- $0.50 / $3.00
- Max output
- 131K
- Knows up to
- —
- Released
- 14 Aug 2026
A 27B dense vision-language model built on the Qwen3.5 architecture and the Qwen3.6-27B predecessor, improving coding and office work in both text and vision. In Qwen's tables it beats Qwen3.6-27B everywhere and matches or beats Qwen3.7-Plus on most agent tasks, under Apache 2.0. model card
qwen3.8-27b- Input / output per 1M
- $0.50 / $3.00
- Context window
- 1M
- Max output
- 131K
- Knows up to
- —
- Reasoning
- Hybrid (thinking on by default; reasoning_effort xhigh / medium / low on the weights)
- Weights
- Open · Apache 2.0
About this card. A model card is the maker's own account of a model: what it's for, how it was trained and tested, and where it falls short. This page condenses Qwen Team, Alibaba's Qwen3.8-27B model card (Hugging Face). Every claim carries its source page.