阿里通义千问团队近日开源全新多模态MoE模型Qwen3.8-Flash-Next，作为Qwen4架构早期预览版。该模型采用高度稀疏的混合专家设计，总参数1250亿，另搭配510亿参数N-gram嵌入表和40亿参数多token预测模块，以极低计算开销实现越级性能表现。
阿里千问发布多模态MoE模型Qwen3.8-Flash，并开源下一代架构Qwen3.8-Flash-Next（Qwen4雏形）。模型总参数125B，搭载51B N-gram Embedding，每token仅激活6B；原生支持26万token，可扩展至100万。训练成本为前代1/9，API定价输入每百万Tokens 1元、输出3元。
DeepSeek团队推出Engram模块，为稀疏大语言模型引入“条件记忆轴”，旨在解决传统Transformer处理重复知识时计算资源浪费的问题。该模块作为混合专家模型的补充，将N-gram嵌入技术融入模型，提升处理重复模式的效率。
nghuyong
ERNIE-Gram是一个通过显式N-Gram掩码语言建模进行预训练的自然语言理解模型
mfleck
基于facebook/wav2vec2-xls-r-300m在Common Voice德语数据集上微调的语音识别模型，集成n-gram语言模型，词错误率8.8%
imvladikon
基于facebook/wav2vec2-xls-r-300m微调的希伯来语语音识别模型，添加了n-gram语言模型增强
%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
VIP会员
