ruby : omit test_build_options locally (#3132)

examples : add HEAPU8 to all of the exported runtime methods (#3134)

wasm : add note about worker.js file generation [no ci] (#3133)

whisper : deprecate WHISPER_CCACHE CMake option (#3131)

stream.wasm : add HEAPU8 to exported runtime methods (#3130)

sync : ggml

cuda : remove nrows_x in mul_mat_q_process_tile (llama/13325)

CUDA: mix virt/real CUDA archs for GGML_NATIVE=OFF (llama/13135)

SYCL: Disable reorder optimize by default and stop setting tensor extras when optimize is disabled (llama/13254)

CUDA: fix bad asserts for partial offload (llama/13337)

CUDA: fix --split-mode row for MMQ (llama/13323)

CUDA: fix logic for clearing padding with -ngl 0 (llama/13320)

SYCL: Disable mul_mat kernels for noncontiguous tensor b (llama/13308)

rpc : use backend registry, support dl backends (llama/13304)

ggml : activate s390x simd for Q3_K (llama/13301)

CUDA: fix race condition in MMQ stream-k fixup (llama/13299)

CUDA: fix race condition in MMQ ids_dst (llama/13294)

vulkan: Additional type support for unary, binary, and copy (llama/13266)

ci : add bindings-java jar artifact to release (#3126)

cli : avoid std::exchange

sync : ggml

vulkan : fix lint (llama/0)

ggml : Enable MMA for BF16 in llamafile_sgemm (llama/13148)

rpc : avoid uninitialized memory in serialize_tensor (llama/13210)

ggml: Don't assert fail when tensor data changes (llama/13222)

build : fix build info on windows (llama/13239)

vulkan: Add bfloat16 support (llama/12554)

vulkan: Handle src1 batch dimension in non-contiguous mat-vec-mul shader (llama/13191)

vulkan : kernels for depthwise 2D convolution (CONV_2D_DW) (ggml/1204)

ci : zip windows artifacts for release uploading (#3124)

ci : add zip extension to xcframework artifact name (#3120)

whisper: remove MSVC warnings pragmas (#3090)

server: update abort mechanism to handle HTTP connection closure (#3112)

cli : support "-" for stdout like stdin (#3050)

docs : Update cli documentation (#3102)

cmake : removed stdc++fs (#3097)

server : update httplib.h to version 0.20.0 (#3101)

ruby : refine HTTP cache feature (#3109)

talk-llama : sync llama.cpp

sync : ggml

CUDA: batched+noncont MMQ, refactor bs>1 MoE code (llama/13199)

vulkan: use uint array index to avoid glslang bug (llama/13193)

ggml : fix ppc64le build (llama/13176)

feat(ggml-cpu): enable z17 compile (llama/13182)

CUDA: fix non-cont. inputs for batched mat mul (llama/13155)

fix(rpc): Improve input validation and error handling (llama/13069)

SYCL: Add all missing unary kernels (llama/13074)

musa: fix typo in cc control (llama/13144)

CUDA: fix q_nope_absorbed prec for DS 2 Lite f16 (llama/13137)

musa: fix build warning (llama/13129)