ggml : mul_mat_id use the same tensor for all the experts (llama/6387)

sync : ggml (#2001)

metal : build metallib + fix embed path (llama/6015)

llama : add pipeline parallelism support (llama/6017)

ggml : reuse quantum structs across backends (llama/5943)

metal : move mm_id indices to shared mem (llama/5982)

ggml : introduce ggml_status (ggml/750)

add some new ops, fix some operators and add batch operations to certain operators. (ggml/747)

IQ4_XS: a 4.25 bpw quantization (llama/5747)

Adding IQ2_S and IQ2_M to complete coverage of the 2-3 bit quantization range (llama/5721)

code : normalize enum names (llama/5697)

IQ3_S: a much better alternative to Q3_K (llama/5676)

Introduce backend GUIDs (ggml/743)

sync : llama.cpp (ggml/0)

1.5 bit quantization (llama/5453)

ggml : add ALiBi support for ggml_soft_max_ext (llama/5488)

ci : add an option to fail on compile warning (llama/3952)

metal : use autoreleasepool to avoid memory leaks (llama/5437)

metal : option to embed MSL source into compiled binary (#1842)

metal : add im2col F32 dst support (llama/5132)

SOTA 3-bit quants (llama/5196)

ggml : add max buffer sizes to opencl and metal backends (llama/5181)

metal : free metal objects (llama/5161)

ci : fix yolo URLs + fix metal capture (ggml/712)

metal : add debug capture backend function (ggml/694)

ggml : add Vulkan backend (llama/2059)

metal : remove unused `n_buffers` and `buffers` (llama/5129)

metal : show compile log messages

metal : disable support for MUL_MAT F32 x F16

ggml : sync ggml-metal.m

metal : create autorelease pool during library build (llama/4970)

metal : log `recommendedMaxWorkingSetSize` on iOS 16+ (llama/4936)

ggml : introduce GGML_CALL function annotation (llama/4850)

metal : correctly set SIMD support flags on iOS (llama/4923)

metal : remove old API (llama/4919)

metal : disable log for loaded kernels (llama/4794)

metal : refactor kernel loading code (llama/4794)

llama : ggml-backend integration (llama/4766)

ggml : SOTA 2-bit quants (add IQ2_XS) (llama/4856)

metal : put encoder debug group behind a define (llama/4873)

metal : fix deprecation warning (ggml/690)

metal : wrap each operation in debug group (ggml/690)

SOTA 2-bit quants (llama/4773)

metal : switch back to default.metallib (ggml/681)

ggml : add error handling to graph_compute (#1714)

metal : add kernel_get_rows_i32

metal : optimize ggml_mul_mat_id (faster Mixtral PP) (llama/4725)

metal : enable shader debugging (cmake option) (llama/4705)

sync : ggml (ggml_scale, ggml_row_size, etc.) (#1677)

sync : ggml (Metal fixes, new ops, tests) (#1633)