AI model · Qwen
Qwen3.5-Flash
Model ID: qwen/qwen3.5-flash-02-23
- Input
- $0.07 per 1M tokens
- Output
- $0.26 per 1M tokens
- Context
- 1M tokens at once
- Max output
- 66K tokens per reply
- Released
- 25 Feb 2026 Qwen
What it can do
- YesReasoning
- YesTool calling
- YesImage input
- NoOpen weights
- NoFree to use
Takes text, image, video; gives text.
How its listing describes it
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...
Sources
- Qwen: Qwen3.5-Flash on OpenRouter — read 2 Oct 2026