ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-17 14:28:07 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

ggml : sync latest ggml repo

Browse Source 
- new Q4 and Q8 quantization
- updated CUDA
This commit is contained in:
Georgi Gerganov
2023-05-20 18:56:30 +03:00
parent bc89f285d8
commit e410cfc3ce
5 changed files with 502 additions and 295 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml-cuda.h
View File

@ -6,6 +6,7 @@ extern "C" {
void ggml_init_cublas(void);
void ggml_cuda_mul(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst);
bool ggml_cuda_can_mul_mat(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst);
size_t ggml_cuda_mul_mat_get_wsize(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst);
void ggml_cuda_mul_mat(const struct ggml_tensor * src0, const struct ggml_tensor * src1, struct ggml_tensor * dst, void * wdata, size_t wsize);
@ -15,6 +16,7 @@ void * ggml_cuda_host_malloc(size_t size);
void ggml_cuda_host_free(void * ptr);
void ggml_cuda_transform_tensor(struct ggml_tensor * tensor);
void ggml_cuda_load_data(const char * fname, struct ggml_tensor * tensors, size_t offset);
#ifdef __cplusplus
}