llama.cpp
cc04d420
- llama: use map for scale tensors instead of manual handling
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 day ago
llama: use map for scale tensors instead of manual handling
References
#28898 - ggml: support quant scales for fp8 and nvfp4
#29791 - vulkan: add fp8 and scaled matmul support
Author
0cc4m
Committer
ORippler
Parents
2d0c9a04
Loading