sipp-sys 0.1.0

Native llama.cpp FFI layer for Sipp
1
2
3
4
5
#include "common.cuh"

#define CUDA_COUNT_EQUAL_CHUNK_SIZE 128

void ggml_cuda_count_equal(ggml_backend_cuda_context & ctx, ggml_tensor * dst);