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

ggml : add Q4_3 quantization (#1082)

Browse source
This commit is contained in:
Georgi Gerganov 2023-04-20 20:35:53 +03:00 • committed by GitHub
parent 6a9661ea5a
commit e0305ead3a
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 293 additions and 37 deletions
Download patch file Download diff file Expand all files Collapse all files

1
llama.h
View file

@ -73,6 +73,7 @@ extern "C" {
LLAMA_FTYPE_MOSTLY_Q4_1 = 3, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q4_1_SOME_F16 = 4, // tok_embeddings.weight and output.weight are F16
LLAMA_FTYPE_MOSTLY_Q4_2 = 5, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q4_3 = 6, // except 1d tensors
};
LLAMA_API struct llama_context_params llama_context_default_params();