vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : fix llama_split_mode enum values in main_gpu document

Browse source 
LLAMA_SPLIT_* were renamed to LLAMA_SPLIT_MODE_* in #5697.
This commit is contained in:
Sutou Kouhei 2024-08-16 19:23:14 +09:00
parent c8ddce8560
commit 68655e3999
1 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

6
include/llama.h
View file

@ -267,9 +267,9 @@ extern "C" {
enum llama_split_mode split_mode; // how to split the model across multiple GPUs
// main_gpu interpretation depends on split_mode:
// LLAMA_SPLIT_NONE: the GPU that is used for the entire model
// LLAMA_SPLIT_ROW: the GPU that is used for small tensors and intermediate results
// LLAMA_SPLIT_LAYER: ignored
// LLAMA_SPLIT_MODE_NONE: the GPU that is used for the entire model
// LLAMA_SPLIT_MODE_ROW: the GPU that is used for small tensors and intermediate results
// LLAMA_SPLIT_MODE_LAYER: ignored
int32_t main_gpu;
// proportion of the model (layers or rows) to offload to each GPU, size: llama_max_devices()