mollysama's picture
llama: Add support for RWKV v7 architecture (llama/12412)
727de7e
raw
history blame
268 Bytes
#ifndef GGML_SYCL_WKV_HPP
#define GGML_SYCL_WKV_HPP
#include "common.hpp"
void ggml_sycl_op_rwkv_wkv6(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
void ggml_sycl_op_rwkv_wkv7(ggml_backend_sycl_context & ctx, ggml_tensor * dst);
#endif // GGML_SYCL_WKV_HPP