Qwen: Qwen3.5-Flash Multimodal
By Qwen
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...
OpenRouter rank
#5
0.02% share
No ranking history yet
This model is in our catalog but hasn't appeared on a tracked leaderboard yet. Trend lines appear after the next ingestion cycle (within 6 hours).