| .. |
|
ggml-blas
|
d337252acf
cmake : Fix broken CMake error messages (ggml/1252)
|
7 months ago |
|
ggml-cann
|
2860d479b4
docker : add cann build pipline (#14591)
|
5 months ago |
|
ggml-cpu
|
baad94885d
ggml : Q2k interleaving implementation - x86/x64 SIMD (#14373)
|
5 months ago |
|
ggml-cuda
|
03d4698218
CUDA: use mma FA kernel for gqa > 4 on RTX 4000 (#15035)
|
5 months ago |
|
ggml-hip
|
b77d11179d
HIP: add GGML_HIP_MMQ_MFMA option to allow disableing the MFMA path. (#14930)
|
5 months ago |
|
ggml-metal
|
793c0d7f46
metal: SSM_SCAN performance (#14743)
|
6 months ago |
|
ggml-musa
|
3f4fc97f1d
musa: upgrade musa sdk to rc4.2.0 (#14498)
|
6 months ago |
|
ggml-opencl
|
5c0eb5ef54
opencl: fix adreno compiler detection logic (#15029)
|
5 months ago |
|
ggml-rpc
|
64bf1c3744
rpc : check for null buffers in get/set/copy tensor endpoints (#14868)
|
6 months ago |
|
ggml-sycl
|
15e92fd337
cuda, sycl : fix batched gemm when ne02 == 1 && ne03 > 1 (#15038)
|
5 months ago |
|
ggml-vulkan
|
5aa1105da2
vulkan: fix build when using glslang that does not support coopmat2 (#15062)
|
5 months ago |
|
ggml-webgpu
|
587d0118f5
ggml: WebGPU backend host improvements and style fixing (#14978)
|
5 months ago |
|
CMakeLists.txt
|
41613437ff
cmake: Add GGML_BACKEND_DIR option (#15074)
|
5 months ago |
|
ggml-alloc.c
|
bf9087f59a
metal : fuse add, mul + add tests (#14596)
|
6 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
|
41613437ff
cmake: Add GGML_BACKEND_DIR option (#15074)
|
5 months ago |
|
ggml-backend.cpp
|
c12bbde372
sched : fix multiple evaluations of the same graph with pipeline parallelism (#14855)
|
6 months ago |
|
ggml-common.h
|
f470bc36be
ggml-cpu : split arch-specific implementations (#13892)
|
7 months ago |
|
ggml-impl.h
|
bf9087f59a
metal : fuse add, mul + add tests (#14596)
|
6 months ago |
|
ggml-opt.cpp
|
6c35981a64
mnist: fix segmentation fault (ggml/1227)
|
8 months ago |
|
ggml-quants.c
|
497be7c01d
ggml-quants : rename best_mad to best_error (ggml/1283)
|
6 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
|
e2b7621e7c
ggml : remove invalid portPos specifiers from dot files (#14838)
|
6 months ago |
|
ggml.cpp
|
fedf034a98
ggml : Print backtrace on uncaught C++ exceptions (ggml/1232)
|
7 months ago |
|
gguf.cpp
|
26a48ad699
ggml : prevent integer overflow in gguf tensor size calculation (#14595)
|
6 months ago |
Christian Kastner
41613437ff
cmake: Add GGML_BACKEND_DIR option (#15074)