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

ggml-quants : 1.625 bpw ternary packing for BitNet 1.58b

Browse source
This commit is contained in:
Francis Couture-Harpin 2024-06-19 12:21:08 -04:00
parent ac146628e4
commit bd807499f7
11 changed files with 594 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
include/llama.h
View file

@ -158,6 +158,8 @@ extern "C" {
LLAMA_FTYPE_MOSTLY_IQ4_XS = 30, // except 1d tensors
LLAMA_FTYPE_MOSTLY_IQ1_M = 31, // except 1d tensors
LLAMA_FTYPE_MOSTLY_BF16 = 32, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q2_2 = 33, // except 1d tensors
LLAMA_FTYPE_MOSTLY_Q1_3 = 34, // except 1d tensors
LLAMA_FTYPE_GUESSED = 1024, // not specified in the model file
};