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

cuda : refactor into multiple files (#6269)

Browse source
This commit is contained in:
slaren 2024-03-25 13:50:23 +01:00 • committed by GitHub
parent ad3a0505e3
commit ae1f211ce2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
59 changed files with 9154 additions and 8987 deletions
Download patch file Download diff file Expand all files Collapse all files

5
ggml-cuda/quantize.cuh Normal file
View file

@ -0,0 +1,5 @@
#include "common.cuh"
#define CUDA_QUANTIZE_BLOCK_SIZE 256
void quantize_row_q8_1_cuda(const float * x, void * vy, const int kx, const int ky, const int kx_padded, cudaStream_t stream);