vrr/prima.cpp
2
0
Fork 0
mirror of https://github.com/Lizonghang/prima.cpp.git synced 2025-09-06 21:59:03 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

add api llama_model_set_n_gpu_layers

Browse source
This commit is contained in:
Lizonghang 2025-01-15 10:47:53 +04:00
parent 9279a2e3ff
commit 1e1ba5bb91
2 changed files with 9 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
common/common.cpp
View file

@ -1440,6 +1440,7 @@ struct llama_init_result llama_init_from_gpt_params(gpt_params & params) {
params.n_gpu_layers = n_gpu_layers[my_rank];
cparams.n_gpu_layers = n_gpu_layers[my_rank];
mparams.n_gpu_layers = n_gpu_layers[my_rank];
llama_model_set_n_gpu_layers(model, n_gpu_layers[my_rank]);
#ifdef LLAMA_DEBUG
device_print_props(dev_info_set, n_world, model, cparams);

8
src/llama.cpp
View file

@ -20705,6 +20705,14 @@ uint32_t llama_model_n_layers(const struct llama_model * model) {
return model->hparams.n_layer;
}
uint32_t llama_model_n_gpu_layers(const struct llama_model * model) {
return model->n_gpu_layers;
}
void llama_model_set_n_gpu_layers(struct llama_model * model, uint32_t value) {
model->n_gpu_layers = value;
}
uint64_t llama_model_n_params(const struct llama_model * model) {
uint64_t nparams = 0;
for (const auto & it : model->tensors_by_name) {