tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-25 17:48:49 -04:00
Code Issues Packages Projects Releases Wiki Activity

CUDA full GPU acceleration, KV cache in VRAM (#1827)

Browse Source 
* Fixed CUDA RoPE

* ggml_cuda_mul_mat_vec_p021

* ggml_cuda_scale

* ggml_cuda_diag_mask_inf

* ggml_is_permuted

* ggml_cuda_cpy

* flatten rows for ggml_cuda_op

* Added a --low-vram option

* Fixed Windows performance

* Fixed LLAMA_CUDA_DMMV_Y > 1 for WizardLM
This commit is contained in:
Johannes Gäßler
2023-06-14 19:47:19 +02:00
committed by GitHub
parent 9254920265
commit 254a7a7a5f
11 changed files with 853 additions and 149 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml-cuda.h
View File

@@ -28,8 +28,10 @@ void ggml_cuda_transform_tensor(void * data, struct ggml_tensor * tensor);
void ggml_cuda_free_data(struct ggml_tensor * tensor);
void ggml_cuda_assign_buffers(struct ggml_tensor * tensor);
void ggml_cuda_assign_buffers_no_scratch(struct ggml_tensor * tensor);
void ggml_cuda_set_main_device(int main_device);
void ggml_cuda_set_scratch_size(size_t scratch_size);
void ggml_cuda_free_scratch(void);
bool ggml_cuda_compute_forward(struct ggml_compute_params * params, struct ggml_tensor * tensor);
#ifdef __cplusplus