CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-30 08:42:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

cuda : re-add q4_0

Browse Source
This commit is contained in:
Georgi Gerganov
2024-08-14 13:37:03 +03:00
parent 503983a69a
commit 62d7b6c87f
2 changed files with 8 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-cuda/mmq.cu
View File

@@ -29,9 +29,9 @@ void ggml_cuda_op_mul_mat_q(
const mmq_args args = {src0_dd_i, src1_ddq_i, dst_dd_i, ne00, row_diff, stride00, src1_padded_row_size, src1_ncols, ne11, nrows_dst};
switch (src0->type) {
// case GGML_TYPE_Q4_0:
// mul_mat_q_case<GGML_TYPE_Q4_0>(ctx, args, stream);
// break;
case GGML_TYPE_Q4_0:
mul_mat_q_case<GGML_TYPE_Q4_0>(ctx, args, stream);
break;
// case GGML_TYPE_Q4_1:
// mul_mat_q_case<GGML_TYPE_Q4_1>(ctx, args, stream);
// break;

5
ggml/src/ggml-cuda/template-instances/mmq-instance-q4_0.cu Normal file
View File

@@ -0,0 +1,5 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../mmq.cuh"
DECL_MMQ_CASE(GGML_TYPE_Q4_0);