| .. |
|
ggml-blas
|
4d3d455d3c
sync : whisper.cpp (ggml/1359)
|
3 months ago |
|
ggml-cann
|
ca709e427b
CANN: add support for partial RoPE and Vision mode (#17543)
|
1 month ago |
|
ggml-cpu
|
79d61896d3
ggml-cpu: add ggml_thread_cpu_relax with Zihintpause support (#17784)
|
1 month ago |
|
ggml-cuda
|
b63509262a
Add DIAG for CUDA (#17873)
|
1 month ago |
|
ggml-hexagon
|
923ae3c619
hexagon: add support for ROPE_NEOX (#17458)
|
2 months ago |
|
ggml-hip
|
80d28f104c
HIP: fix AMDGPU_TARGETS, update documentation (#16803)
|
2 months ago |
|
ggml-metal
|
086a63e3a5
metal: SSM kernel improvements (#17876)
|
1 month ago |
|
ggml-musa
|
11f0af5504
CUDA: faster tile FA, add oob checks, more HSs (#16492)
|
3 months ago |
|
ggml-opencl
|
09c7c50e64
ggml : add circular tiling support to pad, for Vulkan, CUDA, and CPU (used for making seamless textures) (#16985)
|
1 month ago |
|
ggml-rpc
|
7b43f55753
ggml : improve error handling for search path existence checks (#17653)
|
1 month ago |
|
ggml-sycl
|
2e9eab80c2
fix softmax for iGPU (#17838)
|
1 month ago |
|
ggml-vulkan
|
08f9d3cc1d
Vulkan: improve mul_mat_vec_iq1_m (#16907)
|
1 month ago |
|
ggml-webgpu
|
fd57b24c0f
ggml webgpu: unary op suppport, code refactoring, ops support (#17764)
|
1 month ago |
|
ggml-zdnn
|
264f1b5187
zdnn: refactor codebase + add docs (#16178)
|
4 months ago |
|
ggml-zendnn
|
017761daf5
ggml-zendnn : add ZenDNN backend for AMD CPUs (#17690)
|
1 month ago |
|
CMakeLists.txt
|
017761daf5
ggml-zendnn : add ZenDNN backend for AMD CPUs (#17690)
|
1 month ago |
|
ggml-alloc.c
|
86a3f0fad8
ggml : allow fill node alloc inplace (#17870)
|
1 month ago |
|
ggml-backend-impl.h
|
898acba681
rpc : add support for multiple devices (#16276)
|
3 months ago |
|
ggml-backend-reg.cpp
|
017761daf5
ggml-zendnn : add ZenDNN backend for AMD CPUs (#17690)
|
1 month ago |
|
ggml-backend.cpp
|
7f3a72a8ed
ggml : remove redundant n_copies check when setting input/output (#17612)
|
1 month ago |
|
ggml-common.h
|
fd1234cb46
llama : add gpt-oss (#15091)
|
5 months ago |
|
ggml-impl.h
|
389ac78b26
ggml : add ops SOFTPLUS, EXPM1, TRI, SOLVE_TRI, CUMSUM (#17063)
|
2 months ago |
|
ggml-opt.cpp
|
5cdb27e091
finetune: SGD optimizer, more CLI args (#13873)
|
5 months ago |
|
ggml-quants.c
|
f6b4af3d04
ggml : fix uninitialized is_on_grid in quantize_row_iq3_xxs_impl (#15928)
|
4 months ago |
|
ggml-quants.h
|
fd1234cb46
llama : add gpt-oss (#15091)
|
5 months 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
|
02e409a5be
ggml : Provide macos-specific backtrace printing to avoid terminal death (#17869)
|
1 month ago |
|
ggml.cpp
|
fedf034a98
ggml : Print backtrace on uncaught C++ exceptions (ggml/1232)
|
7 months ago |
|
gguf.cpp
|
37adc9c6ba
ggml, llama : use defaulted constructors/destructors (#17649)
|
1 month ago |
Neo Zhang Jianyu
2e9eab80c2
fix softmax for iGPU (#17838)