llama.cpp
c02faf8f
- reduce shmem use through buffer reuse
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
8 days ago
reduce shmem use through buffer reuse
References
#31323 - vulkan: optimize prompt processing Flash Attention for RDNA3 and RDNA4
Author
0cc4m
Committer
0cc4m
Parents
075deb79
Loading