mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-27 03:44:35 +00:00
2f3c1466ff
* llava: Add ACC OP for GPU acceleration to the Vulkan backend in the LLAVA CLIP model. - The CLIP model now prioritizes the Vulkan backend over the CPU when vulkan available. - A GGML_OP_ACC shader has been added. - The encoding performance of the CLIP model improved from 4.2s on the CPU to 0.9s on the GPU. Signed-off-by: Changyeon Kim <cyzero.kim@samsung.com> * fix-up coding style. Signed-off-by: Changyeon Kim <cyzero.kim@samsung.com> * Fix-up the missing initial parameter to resolve the compilation warning. Signed-off-by: Changyeon Kim <cyzero.kim@samsung.com> * [fix] Add missing parameters. Signed-off-by: Changyeon Kim <cyzero.kim@samsung.com> * [fix] Use nb1 and nb2 for dst. Signed-off-by: Changyeon Kim <cyzero.kim@samsung.com> * Fix check results ggml_acc call --------- Signed-off-by: Changyeon Kim <cyzero.kim@samsung.com> Co-authored-by: 0cc4m <picard12@live.de> |
||
|---|---|---|
| .. | ||
| ggml-cann | ||
| ggml-cuda | ||
| ggml-sycl | ||
| kompute@4565194ed7 | ||
| kompute-shaders | ||
| llamafile | ||
| vulkan-shaders | ||
| CMakeLists.txt | ||
| ggml-aarch64.c | ||
| ggml-aarch64.h | ||
| ggml-alloc.c | ||
| ggml-backend-impl.h | ||
| ggml-backend.c | ||
| ggml-blas.cpp | ||
| ggml-cann.cpp | ||
| ggml-common.h | ||
| ggml-cuda.cu | ||
| ggml-impl.h | ||
| ggml-kompute.cpp | ||
| ggml-metal.m | ||
| ggml-metal.metal | ||
| ggml-quants.c | ||
| ggml-quants.h | ||
| ggml-rpc.cpp | ||
| ggml-sycl.cpp | ||
| ggml-vulkan.cpp | ||
| ggml.c | ||