vrr/prima.cpp
2
0
Fork 0
mirror of https://github.com/Lizonghang/prima.cpp.git synced 2025-09-10 08:34:33 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

add memory copy speed test

Browse source
This commit is contained in:
Lizonghang 2024-12-09 10:07:42 +04:00
parent 1aee5bd6da
commit d78fa427e7
4 changed files with 179 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

6
include/llama.h
View file

@ -528,6 +528,12 @@ extern "C" {
// Returns the total number of parameters in the model
LLAMA_API uint64_t llama_model_n_params(const struct llama_model * model);
// Returns the embedding size of K in grouped query attention
LLAMA_API uint32_t llama_model_n_embd_k_gqa(struct llama_model * model);
// Returns the embedding size of V in grouped query attention
LLAMA_API uint32_t llama_model_n_embd_v_gqa(struct llama_model * model);
// Return the size of compute buffer size, including input tensors and activations
LLAMA_API void llama_model_compute_buf_size(
uint64_t * cpu_buf,