vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-11 01:24:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge: Testing speed of tensor cores vs MMQ

Browse source
This commit is contained in:
Concedo 2023-10-28 16:41:42 +08:00
commit 2ea3b567cf
5 changed files with 154 additions and 28 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama.cpp
View file

@ -1583,12 +1583,14 @@ static void llama_kv_cache_seq_shift(
enum llama_fver {
GGUF_FILE_VERSION_V1 = 1,
GGUF_FILE_VERSION_V2 = 2,
GGUF_FILE_VERSION_V3 = 3,
};
static const char * llama_file_version_name(llama_fver version) {
switch (version) {
case GGUF_FILE_VERSION_V1: return "GGUF V1 (support until nov 2023)";
case GGUF_FILE_VERSION_V2: return "GGUF V2 (latest)";
case GGUF_FILE_VERSION_V2: return "GGUF V2";
case GGUF_FILE_VERSION_V3: return "GGUF V3 (latest)";
}
return "unknown";