| .. |
|
ggml-cann
|
904837e0cb
cann: fix crash when llama-bench is running on multiple cann devices (#9627)
|
vor 1 Jahr |
|
ggml-cuda
|
aaa4099925
CUDA: remove bad assert (ggml/972)
|
vor 1 Jahr |
|
ggml-sycl
|
e62e9789cd
Revert "[SYCL] fallback mmvq (#9088)" (#9579)
|
vor 1 Jahr |
|
kompute @ 4565194ed7
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
vor 1 Jahr |
|
kompute-shaders
|
06943a69f6
ggml : move rope type enum to ggml.h (#8949)
|
vor 1 Jahr |
|
llamafile
|
23e0d70bac
ggml : move common CPU backend impl to new header (#9509)
|
vor 1 Jahr |
|
vulkan-shaders
|
544f409b4b
vulkan : argsort barriers must be under uniform control flow (ggml/951)
|
vor 1 Jahr |
|
CMakeLists.txt
|
70392f1f81
ggml : add AVX512DQ requirement for AVX512 builds (#9622)
|
vor 1 Jahr |
|
ggml-aarch64.c
|
6a0f779484
ggml : add run-time detection of neon, i8mm and sve (#9331)
|
vor 1 Jahr |
|
ggml-aarch64.h
|
370b1f7e7a
ggml : minor naming changes (#8433)
|
vor 1 Jahr |
|
ggml-alloc.c
|
d09770cae7
ggml-alloc : fix list of allocated tensors with GGML_ALLOCATOR_DEBUG (#9573)
|
vor 1 Jahr |
|
ggml-backend-impl.h
|
424c5d00a9
ggml/examples: add backend support for numerical optimization (ggml/949)
|
vor 1 Jahr |
|
ggml-backend.c
|
27609c49b9
ggml : fix trailing whitespace (#0)
|
vor 1 Jahr |
|
ggml-blas.cpp
|
d6a04f872d
ggml : hide ggml_object, ggml_cgraph, ggml_hash_set (#9408)
|
vor 1 Jahr |
|
ggml-cann.cpp
|
424c5d00a9
ggml/examples: add backend support for numerical optimization (ggml/949)
|
vor 1 Jahr |
|
ggml-common.h
|
9bc6db28d0
ggml-quants : ternary packing for TriLMs and BitNet b1.58 (#8151)
|
vor 1 Jahr |
|
ggml-cpu-impl.h
|
23e0d70bac
ggml : move common CPU backend impl to new header (#9509)
|
vor 1 Jahr |
|
ggml-cuda.cu
|
7691654c68
mtgpu: enable VMM (#9597)
|
vor 1 Jahr |
|
ggml-impl.h
|
23e0d70bac
ggml : move common CPU backend impl to new header (#9509)
|
vor 1 Jahr |
|
ggml-kompute.cpp
|
424c5d00a9
ggml/examples: add backend support for numerical optimization (ggml/949)
|
vor 1 Jahr |
|
ggml-metal.m
|
cad341d889
metal : reduce command encoding overhead (#9698)
|
vor 1 Jahr |
|
ggml-metal.metal
|
bf9c1013ac
metal : use F32 prec for K*Q in vec FA (#9595)
|
vor 1 Jahr |
|
ggml-quants.c
|
6a0f779484
ggml : add run-time detection of neon, i8mm and sve (#9331)
|
vor 1 Jahr |
|
ggml-quants.h
|
6a0f779484
ggml : add run-time detection of neon, i8mm and sve (#9331)
|
vor 1 Jahr |
|
ggml-rpc.cpp
|
424c5d00a9
ggml/examples: add backend support for numerical optimization (ggml/949)
|
vor 1 Jahr |
|
ggml-sycl.cpp
|
e62e9789cd
Revert "[SYCL] fallback mmvq (#9088)" (#9579)
|
vor 1 Jahr |
|
ggml-vulkan.cpp
|
00b7317e63
vulkan : do not use tensor->extra (#9407)
|
vor 1 Jahr |
|
ggml.c
|
e98c1c188e
test: fix OPT_STEP_ADAMW for test-backend-ops (ggml/974)
|
vor 1 Jahr |
Radoslav Gerganov
00b7317e63
vulkan : do not use tensor->extra (#9407)