CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add enum llama_ftype, sync ggml_type to model files (#709)

Browse Source
This commit is contained in:
Stephan Walter
2023-04-11 15:03:51 +00:00
committed by GitHub
parent 2663d2c678
commit 3e6e70d8e8
5 changed files with 74 additions and 57 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
llama.h
View File

@@ -65,6 +65,14 @@ extern "C" {
void * progress_callback_user_data;
};
// model file types
enum llama_ftype {
LLAMA_FTYPE_ALL_F32 = 0,
LLAMA_FTYPE_MOSTLY_F16 = 1, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q4_0 = 2, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q4_1 = 3, // except 1d tensors
};
LLAMA_API struct llama_context_params llama_context_default_params();
LLAMA_API bool llama_mmap_supported();
@@ -85,7 +93,7 @@ extern "C" {
LLAMA_API int llama_model_quantize(
const char * fname_inp,
const char * fname_out,
int itype);
enum llama_ftype ftype);
// Returns the KV cache that will contain the context for the
// ongoing prediction with the model.