ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-27 18:01:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

q2_k_r4: Zen4

Browse Source 
PP-512(LLaMA-3.1-8B) = 256 t/s
This commit is contained in:
Iwan Kawrakow
2024-12-11 16:28:56 +02:00
parent 9469af87f7
commit 93c6e295ee
10 changed files with 263 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/include/ggml.h
View File

@@ -412,6 +412,7 @@ extern "C" {
GGML_TYPE_Q4_0_R4 = 202,
GGML_TYPE_Q5_0_R4 = 206,
GGML_TYPE_Q8_0_R4 = 208,
GGML_TYPE_Q2_K_R4 = 210,
GGML_TYPE_Q3_K_R4 = 211,
GGML_TYPE_Q4_K_R4 = 212,
GGML_TYPE_Q5_K_R4 = 213,
@@ -482,6 +483,7 @@ extern "C" {
GGML_FTYPE_MOSTLY_Q4_0_R4 = 202, // except 1d tensors
GGML_FTYPE_MOSTLY_Q8_0_R4 = 207, // except 1d tensors
GGML_FTYPE_MOSTLY_Q5_0_R4 = 208, // except 1d tensors
GGML_FTYPE_MOSTLY_Q2_K_R4 = 210, // except 1d tensors
GGML_FTYPE_MOSTLY_Q3_K_R4 = 211, // except 1d tensors
GGML_FTYPE_MOSTLY_Q4_K_R4 = 212, // except 1d tensors
GGML_FTYPE_MOSTLY_Q5_K_R4 = 215, // except 1d tensors