issues
search
likejazz
/
llama3.cuda
llama3.cuda is a pure C/CUDA implementation for Llama 3 model.
MIT License
294
stars
20
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Use CUDA event API for benchmarking
#6
meneraing
opened
1 month ago
0
Output same word
#5
dmuestc
opened
3 months ago
2
2,823 tokens/s seems extremely high!
#4
romitjain
opened
3 months ago
1
OneAPI Version? That's awesome!
#3
ElliottDyson
opened
4 months ago
1
한글 및 utf8 멀티바이트 토크나이저 지원
#2
go-noah
closed
4 months ago
2
Build error
#1
MoffKalast
opened
4 months ago
1