Qwen 3.8 Flash Next locally on simple mobile phone at 3.5 tok/s

查看原文

Qwen模型发布

热度趋势

↓ 降温 17%

最近 24 小时与此前 24 小时对比 · 7 天曲线

百分比基于当前可用热度信号，而非评论数或独立用户人数。

推荐理由

Qwen 相关模型动态已经出现，适合跟踪能力变化、生态影响和后续可用性。

AI 摘要

The Qwen 3.8 Flash Next model, which is 80GB, can now run locally on a mid-range Android phone with 12GB of RAM, achieving 3.5 tokens per second. This performance is attributed to optimizations and low quantization on the dense part of the model. This development demonstrates the feasibility of running such models on affordable mobile devices, specifically on phones costing $400–$500.