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

124 points by mfiguiere 12 hours ago | 10 comments

arikrahman 11 hours ago |
I am very impressed with the KV Cache Compression work as well as the prefix cacheing making queries converge on practically free.
smy20011 10 hours ago |
Removed
sebmellen 10 hours ago |
The writing feels human to me… and I call out AI slop as much as possible.
girvo 10 hours 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 10 hours ago |
you don't know him?
vivzkestrel 10 hours ago |
404 on the blog page? https://zartbot.github.io/blog/
sanufar 9 hours ago |
Yeah, for some reason their /blog/ is 404ing but for anyone interested in their other articles, https://github.com/zartbot/blog/ has all of them, just sub out anything after /blog/.* with the folder path (e.g https://zartbot.github.io/blog/arch/jalapeno/ from https://github.com/zartbot/blog/tree/main/arch/jalapeno)
mmastrac 8 hours ago |
I've been working with an automatic incremental context compactor enabled and it's been surprisingly helpful. It was particularly effective with DS41f - I think I was running at an effective session length of 5M, with the model running around 300k-400k and it was holding on both speed and intelligence.
TBH I also ran the 400tok/s preview and that was just nuts. I just let the thing compact over and over over the course of a day attacking a couple of tough problems
nchmy 3 hours ago |
Can you share a link to thr automatic compactor? I've been noticing that when I get to around 60% context window, the cache will simply break and suddenly I've paid 50x more than expected. The only solution seems to be to compact or start a new session.
mmastrac 14 minutes ago |
Send me an email- it's not public just yet