ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-05-01 03:41:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

iq2_k: Basics

Browse Source 
Quantize/dequantize, CUDA deqantize, AVX512 iqk_mul_mat.
This commit is contained in:
Iwan Kawrakow
2024-07-28 19:43:18 +03:00
committed by Kawrakow
parent 291066e6df
commit c85e139c68
15 changed files with 451 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/include/ggml.h
View File

@@ -390,6 +390,7 @@ extern "C" {
GGML_TYPE_IQ2_BN = 35,
GGML_TYPE_Q8_K64 = 36,
GGML_TYPE_IQ4_K = 37,
GGML_TYPE_IQ2_K = 38,
GGML_TYPE_COUNT,
};
@@ -437,6 +438,7 @@ extern "C" {
GGML_FTYPE_MOSTLY_IQ1_BN = 28, // except 1d tensors
GGML_FTYPE_MOSTLY_IQ2_BN = 29, // except 1d tensors
GGML_FTYPE_MOSTLY_IQ4_K = 30, // except 1d tensors
GGML_FTYPE_MOSTLY_IQ2_K = 31, // except 1d tensors
};
// available tensor operations: