Datasets:

Modalities:
Text
Formats:
text
Size:
< 1K
ArXiv:
Libraries:
Datasets
llama.cpp / ggml /include /ggml-zdnn.h
dlxj
todo: 基于 CUDA 13.0 编译
2517be1
#pragma once
#include "ggml.h"
#include "ggml-backend.h"
#ifdef __cplusplus
extern "C" {
#endif
// device buffer
GGML_BACKEND_API ggml_backend_buffer_type_t ggml_backend_zdnn_buffer_type(void);
GGML_BACKEND_API ggml_backend_reg_t ggml_backend_zdnn_reg(void);
#ifdef __cplusplus
}
#endif