nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity
6,108 Commits 1 Branch 0 Tags
20638e4f16fcc21f836c7556e83bbf532bb5a0f0
Commit Graph
1 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Shouzheng LiuandGitHub 1cbf561466 metal : new q4_0 matrix-vector kernel (#2188) 
Prefetch data to improve GPU utilization. ~48% faster for 33B model.
 2023-07-12 23:10:55 +03:00