How is this model half the size of DeepSeek V4 Pro? Is it because DeepSeek did m... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

andai 54 days ago | parent | context | favorite | on: GLM-5.2 – How to Run Locally

How is this model half the size of DeepSeek V4 Pro? Is it because DeepSeek did more aggressive cost cutting on the attention mechanism?

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact