目前，3 款模型均已在魔搭社区、Hugging Face 开源上线，同时，我们还一并开源了 Qwen3.5-35B-A3B-Base 基座模型。
This started with Addition Under Pressure, where I gave Claude Code and Codex the same prompt: train the smallest possible transformer that can do 10-digit addition with at least 99% accuracy. Claude Code came back with 6,080 parameters and Codex came back with 1,644. The community has since pushed this dramatically lower.，更多细节参见51吃瓜
，详情可参考搜狗输入法2026
然而，随着数据量的爆发增长，DRAM及NAND在耗电量及数据访问速度上依旧无法跟上需求的脚步。他们在需要高速运算的应用场景中也有一些阻碍。。关于这个话题，旺商聊官方下载提供了深入分析
Екатерина Ештокина