OpenTransformer's picture
Q1_0_g128 CPU kernel fix + AVX2 SIMD (fork of PrismML-Eng/llama.cpp)
03ba2cd verified
#include "common.cuh"
#define CUDA_CPY_BLOCK_SIZE 64
void ggml_cuda_cpy(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, ggml_tensor * src1);
void ggml_cuda_dup(ggml_backend_cuda_context & ctx, ggml_tensor * dst);