このページで分かること
The Qwen3.5 Series 35B-A3B is a native vision-language model designed with a hybrid architecture that integrates linear attention mechanisms and a sparse mixture-of-experts model, achieving higher inference efficiency. I
- Alibaba Cloud · Qwen · qwen/qwen3.5-35b-a3b
- text+image+video->text · 中国モデルルート
- 262,144 context · $0.163 input