@anon
sign up
@anon
sign up
10 sats
\
0 replies
\
@Zepasta
16 Aug 2023
\
on: How Is LLaMa.cpp Possible?
tech
I'm getting around 5 token/sec with i7 + 16GB RAM + RTX 2000 using LLaMa 7B. It's not fast enough for me to consider it usable.
write
preview
reply
100 sats