ggml: initial IBM zDNN backend (llama/14975)

cmake: Add GGML_BACKEND_DIR option (llama/15074)

ggml: Add initial WebGPU backend (llama/14521)

sync : resolve conflicts (#0)

build : suppress gcc15 compile warnings (llama/14261)

ggml-backend : fix backend search path (llama/12330)

ggml-backend : make path_str compatible with C++20 (llama/12269)

ggml : portability fixes for VS 2017 (llama/12150)

ggml-backend : keep paths in native string type when possible (llama/12144)

ggml : allow loading backend with env variable (ggml/1059)

ggml : use wstring for backend search paths (llama/10960)

ggml : fix run-time on FreeBSD in get_executable_path() (llama/10948)

Introducing experimental OpenCL backend with support for Qualcomm Adreno GPUs (llama/10693)

Fix crash caused by ggml_backend_load_all when launching on Android Activity (llama/10812)

ggml: load all backends from a user-provided search path (llama/10699)

ggml : add predefined list of CPU backend variants to build (llama/10626)

ggml : move AMX to the CPU backend (llama/10570)

llama : accept a list of devices to use to offload a model (llama/10497)

ggml : add support for dynamic loading of backends (llama/10469)

ggml : build backends as libraries (llama/10256)