---
name: 'Qwen: Qwen3.5-Flash'
slug: qwen-3-5-flash-02-23
type: Multimodal
providers:
  - openrouter
description: >-
  The Qwen3.5 native vision-language Flash models are built on a hybrid
  architecture that integrates a linear attention mechanism with a sparse
  mixture-of-experts model, achieving higher inference efficiency. Compared to
  the...
tags:
  - multimodal
is_open_source: false
context_window: 1000000
max_output: 65536
pricing:
  - image: 0
    input: 0.065
    output: 0.26
    provider: openrouter
release_date: '2024-01-01'
openrouter_id: qwen/qwen3.5-flash-02-23
capabilities:
  audio: false
  vision: true
  tool_use: false
  citations: false
  reasoning: false
---
Qwen: Qwen3.5-Flash is a Multimodal model.
The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the...