llama.cpp
CUDA: noncont MMVQ + batched bs1 MUL_MAT_ID
#13014
Merged
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Overview
Commits
3
Changes
View On
GitHub
Commits
CUDA: noncont MMVQ + batched bs1 MUL_MAT_ID
JohannesGaessler
committed
1 year ago
fix logic for RoPE support, CUDA graphs
JohannesGaessler
committed
1 year ago
add asserts for memory layout and batch size
JohannesGaessler
committed
1 year ago
Loading
