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. Its overall...
模态
文本图像视频
输入 / 输出价格
₩548 / ₩2,194
1M
上下文
26.21万
发布日期
2026年2月26日
此模型暂无基准测试数据。