llama.cpp
c145db27
- webgpu : use f32 while dequantizing to avoid FTZ (#32140)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 day ago
webgpu : use f32 while dequantizing to avoid FTZ (#32140)
References
#32140 - webgpu : use f32 while dequantizing to avoid FTZ
Author
tokinasin
Parents
ede61067
Loading