llama.cpp
17720fad
- metal : parallel reduce across heads
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 year ago
metal : parallel reduce across heads
References
#5021 - ggml : add Flash Attention
Author
ggerganov
Committer
ggerganov
Parents
77d08f32
Loading