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

CUDA: use MMQ instead of cuBLAS by default (#8075)

Browse source
This commit is contained in:
Johannes Gäßler 2024-06-24 17:43:42 +02:00 • committed by GitHub
parent d62e4aaa02
commit a818f3028d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 124 additions and 122 deletions
Download patch file Download diff file Expand all files Collapse all files

2
ggml-cuda/mmvq.cuh
View file

@ -1,5 +1,7 @@
#include "common.cuh"
#define MMVQ_MAX_BATCH_SIZE 8 // Max. batch size for which to use MMVQ kernels.
void ggml_cuda_op_mul_mat_vec_q(
ggml_backend_cuda_context & ctx,
const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst, const char * src0_dd_i, const float * src1_ddf_i,