搜索
关闭导航

包含标签" mixture of experts (MoE) architecture"的内容

腾讯开源Hunyuan-A13B：动态推理+MoE架构，AIME准确率87.3%超OpenAI o1

腾讯开源Hunyuan-A13B：动态推理+MoE架构，AIME准确率87.3%超OpenAI o1

AI妹 1 个月前 7 0

Tencent has recently announced the open-source release of its new language model "**Hunyuan-A13B**

Alibaba Qwen 87.3% accuracy rate 250 billion STEM tokens mixture of experts (MoE) architecture OpenAI
查看详情
阿里巴巴开源Qwen3-Coder AI编程模型 采用MoE架构性能卓越适配多场景免费可用

阿里巴巴开源Qwen3-Coder AI编程模型 采用MoE架构性能卓越适配多场景免费可用

AI妹 1 个月前 8 0

The morning of July 23, Alibaba officially launched its new open-source AI programming model - Qwe

agent tasks web development tools calling 1M tokens extended context 3.5 billion activated parameters
查看详情
阿里巴巴Qwen3-Coder成全球最受欢迎开源AI编程模型

阿里巴巴Qwen3-Coder成全球最受欢迎开源AI编程模型

AI妹 1 个月前 11 0

On July 24, the world's largest AI open-source community, Hugging Face, released the latest large

Hugging Face open-source AI programming model Claude4 mixture of experts (MoE) architecture Qwen3-Coder
查看详情

资讯姬

文章数量9911 总阅读量90.553k 总评论量1 会员数量2

最新评论

emlog

emlog1 个月前

发表在：欢迎使用emlog

这是系统生成的演示评论

链接

搜索一下

关于我 免责协议 隐私保护 广告合作 侵权删除
AI导航 AI导航
本站由emlog驱动