Qwen: Qwen3.5-Flash
0.0 / 10
1M context
Proprietary
About this 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...
Technical Specs
- Architecture: text+image+video->text
- Context Window: 1,000,000 tokens
- Input Modalities: text
Hardware Requirements
- API-only (no local hardware needed)
Pricing
| Input | Output | Currency |
|---|---|---|
| 0.065 / 1M tokens | 0.26 / 1M tokens | USD |