DeepSeek-v4.1 Flash: Pushing the Limits of KV Cache Compression (zartbot.github.io)

11 comments

[ 2.6 ms ] story [ 33.6 ms ] thread

I am very impressed with the KV Cache Compression work as well as the prefix cacheing making queries converge on practically free.

smy20011 1mo ago ↗

Should we flag this since It's AI generated?

sebmellen 1mo ago ↗

The writing feels human to me… and I call out AI slop as much as possible.

girvo 1mo 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 1mo ago ↗

you don't know him?

vivzkestrel 1mo ago ↗

404 on the blog page? https://zartbot.github.io/blog/

sanufar 1mo 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)

[dead] N_Lens 1mo ago ↗

[dead]

mmastrac 1mo 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 1mo 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 1mo ago ↗

Send me an email- it's not public just yet