| .. |
|
ggml-blas
|
5931c1f233
ggml : add support for dynamic loading of backends (#10469)
|
1 year ago |
|
ggml-cann
|
9bacd6b374
[CANN] get_rows and dup optimization (#12671)
|
9 months ago |
|
ggml-cpu
|
360dc22c00
cpu : rm unused variable (ggml/1166)
|
10 months ago |
|
ggml-cuda
|
250d7953e8
ggml : faster ssm scan (#10558)
|
9 months ago |
|
ggml-hip
|
becade5de7
HIP: implement FlashAttention via rocWMMA for CDNA and RDNA3+ (#12032)
|
10 months ago |
|
ggml-kompute
|
ba1cb19cdd
llama : add Qwen2VL support + multimodal RoPE (#10361)
|
1 year ago |
|
ggml-metal
|
3fd072a540
metal : use F32 prec in FA kernels (#12688)
|
9 months ago |
|
ggml-musa
|
b1b132efcb
cuda : enable CUDA Graph on CUDA Toolkit < 12.x (#12394)
|
10 months ago |
|
ggml-opencl
|
f423981ac8
opencl : fix memory allocation size (#12649)
|
9 months ago |
|
ggml-rpc
|
ab6ab8f809
rpc : send hash when tensor data is above some fixed threshold (#12496)
|
10 months ago |
|
ggml-sycl
|
8293970542
SYCL: Rename oneMKL to oneMath (#12192)
|
9 months ago |
|
ggml-vulkan
|
f01bd02376
vulkan: Implement split_k for coopmat2 flash attention. (#12627)
|
9 months ago |
|
CMakeLists.txt
|
a69f846351
cmake : fix ccache conflict (#12522)
|
10 months ago |
|
ggml-alloc.c
|
70680c48e5
ggml : upgrade init_tensor API to return a ggml_status (#11854)
|
10 months ago |
|
ggml-backend-impl.h
|
70680c48e5
ggml : upgrade init_tensor API to return a ggml_status (#11854)
|
10 months ago |
|
ggml-backend-reg.cpp
|
ba7654380a
ggml-backend : fix backend search path (#12330)
|
10 months ago |
|
ggml-backend.cpp
|
5bbe6a9fe9
ggml : portability fixes for VS 2017 (#12150)
|
10 months ago |
|
ggml-common.h
|
492d7f1ff7
musa: fix all warnings, re-enable `-DLLAMA_FATAL_WARNINGS=ON` in ci and update doc (#12611)
|
10 months ago |
|
ggml-impl.h
|
24feaec057
ggml : riscv: add 128-bit RVV support (#12530)
|
10 months ago |
|
ggml-opt.cpp
|
02e4eaf22f
ggml-opt: fix data corruption (ggml/1022)
|
1 year ago |
|
ggml-quants.c
|
5bbe6a9fe9
ggml : portability fixes for VS 2017 (#12150)
|
10 months ago |
|
ggml-quants.h
|
ae8de6d50a
ggml : build backends as libraries (#10256)
|
1 year ago |
|
ggml-threading.cpp
|
ae8de6d50a
ggml : build backends as libraries (#10256)
|
1 year ago |
|
ggml-threading.h
|
cb13ef85a4
remove CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS (#10797)
|
1 year ago |
|
ggml.c
|
e0e912f49b
llama : add option to override model tensor buffers (#11397)
|
9 months ago |
|
gguf.cpp
|
a6f32f0b34
Fix clang warning in gguf_check_reserved_keys (#12686)
|
9 months ago |
Jeff Bolz
f01bd02376
vulkan: Implement split_k for coopmat2 flash attention. (#12627)