Qwen: Qwen3.5-Flash
Alibaballm
About
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...
Try Qwen: Qwen3.5-Flash
Test this model in the Sandbase Playground with your own prompts.
Open in Playground