//
sign in
Post
by @danabra.mov
PostEmbed
by @danabra.mov
Record
by @jimpick.com
Record
by @atsui.org
+ new component
Post
How is DeepSeek V4 so INSANELY cheap? 🤔 Compared to a GQA baseline, it's new *compressed attention* mechanism (CSA and HCA) slashes the KV cache memory cost by 98% 🤯 at a 1M-token context! Here’s how: youtu.be/q8holiIirgo
1mo
Jia-Bin Huang