DeepSeek-v4.1 Flash: Pushing the Limits of KV Cache Compression

3 comments

• arikrahman 59 minutes ago
  I am very impressed with the KV Cache Compression work as well as the prefix cacheing making queries converge on practically free.
• vivzkestrel 16 minutes ago
  404 on the blog page? https://zartbot.github.io/blog/
• smy20011 46 minutes ago
  Should we flag this since It's AI generated?
  [-]
  • girvo 16 minutes ago
    The website design definitely is, but I don’t know if the content is? This reads pretty human to me and is quite interesting to boot!
  • yunfei 9 minutes ago
    you don't know him?
  • sebmellen 31 minutes ago
    The writing feels human to me… and I call out AI slop as much as possible.