llama-cpp-sys-4 0.7.0

Low Level Bindings to llama.cpp
Documentation 
1
2
3
4
5
6
7

#include "common.cuh"

void ggml_cuda_op_moe_weighted_reduction(ggml_backend_cuda_context & ctx,
                                         const ggml_tensor *         experts,
                                         const ggml_tensor *         expert_scale,
                                         const ggml_tensor *         weights,
                                         ggml_tensor *               dst);