mirror of
https://github.com/LostRuins/koboldcpp.git
synced 2025-09-11 09:34:37 +00:00
d1bb126605
# Conflicts: # README.md # llama.cpp # otherarch/sdcpp/SDCPP_LICENSE # scripts/sync-ggml-am.sh # scripts/sync-ggml.sh
5 lines
200 B
Text
5 lines
200 B
Text
#include "common.cuh"
|
|
|
|
#define CUDA_QUANTIZE_BLOCK_SIZE 256
|
|
|
|
void quantize_row_q8_1_cuda(const float * x, void * vy, const int64_t kx, const int64_t ky, const int64_t kx_padded, cudaStream_t stream);
|