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

 (zartbot.github.io)

52 points | by mfiguiere 2 hours ago

3 comments

  • arikrahman 52 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 9 minutes ago
    404 on the blog page? https://zartbot.github.io/blog/
  • smy20011 39 minutes ago
    Should we flag this since It's AI generated?
    [-]
    • girvo 10 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 2 minutes ago
      you don't know him?
    • sebmellen 24 minutes ago
      The writing feels human to me… and I call out AI slop as much as possible.