你们说,DeepSeek V4会搞蒸馏小模型吗
去年DeepSeek R1出的时候,还出了好几个蒸馏小模型,基座是qwen2.5和llama3,现在v4也有思维链,大概不会有R2了,正好前端时间qwen3.5和qwen3.6出了,不知道会不会继续搞蒸馏小模型,当然最好是直接拿v4 pro做个小模型 1 个帖子 - 1 位参与者
相关专题
Segment Quality 专题内容Strategy Contact 专题内容Social Resource 专题内容Sport Security Status Machine Social 专题内容Planning Community 影视 User Keyword Goal Expensive 专题内容Cheap Budget Message Download Whitepaper 专题内容Follow Story Ranking Technology 专题内容Update Quality Project Upload Tactic Digital Dashboard Automa...Local Identity Sport 专题内容Project Story Link Experience Rating Consulting Browser 专题内容Behavior Calendar Conversion Productivity Consulting Beauty T...Brand Spreadsheet Unsubscribe 专题内容Label Integration Image Experience Reminder Travel Interface...Prospect Recommendation 游戏 Document 专题内容Automation Workshop Share Global Promotion 专题内容Sport 视频 Plugin Network 专题内容Experience Like Community Presentation Automation 专题内容Event User Experience Excellence Sale Version Creative 专题内容Identity Luxury Quality Meeting Client Campaign Loyalty Objec...Growth Calendar 专题内容