vrr/prima.cpp
2
0
Fork 0
mirror of https://github.com/Lizonghang/prima.cpp.git synced 2025-09-10 16:34:35 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

rename arg --keep-inp-out-in-metal to --keep-out-in-metal

Browse source
This commit is contained in:
Lizonghang 2025-01-23 23:17:06 +04:00
parent 5fcf020cfb
commit 1c0087e919
5 changed files with 14 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

4
include/llama.h
View file

@ -312,7 +312,7 @@ extern "C" {
bool use_mmap; // use mmap if possible
bool use_mlock; // force system to keep model in RAM
bool check_tensors; // validate model tensor data
bool keep_inp_out_in_metal; // whether to keep input/output weight in metal
bool keep_out_in_metal; // whether to keep output weights in metal memory
};
// NOTE: changing the default values of parameters marked as [EXPERIMENTAL] may cause crashes or incorrect results in certain configurations
@ -323,7 +323,7 @@ extern "C" {
uint32_t n_layer_window[32];// number of layers to process in each compute
uint32_t n_gpu_layers; // number of layers to process on GPU
bool unload; // whether to unload layer weights after use
bool keep_inp_out_in_metal; // whether to keep input/output weight in metal
bool keep_out_in_metal; // whether to keep output weights in metal memory
char * master_ip; // ip address of the master node
char * next_node_ip; // ip address of the next node
uint32_t n_ctx; // text context, 0 = from model