b4231

build: update Makefile comments for C++ version change (#10598)

b4230

ggml-cpu: replace AArch64 NEON assembly with intrinsics in ggml_gemv_…

b4227

vulkan: Dynamic subgroup size support for Q6_K mat_vec (#10536)

* subgroup 64 version with subgroup add. 15% faster

scalable version

tested for subgroup sizes 16-128

* check for subgroup multiple of 16 and greater than 16

* subgroup sizes are always a power of 2 (https://github.com/KhronosGroup/GLSL/issues/45)

* force 16 sequential threads per block

* make 16 subgroup size a constant

b4226

ggml : move AMX to the CPU backend (#10570)

* ggml : move AMX to the CPU backend

---------

Co-authored-by: Georgi Gerganov <[email protected]>

b4224

imatrix : support combine-only (#10492)

* imatrix-combine-only idea

* ensured that behavior consistent with log

b4222

ggml : fix I8MM Q4_1 scaling factor conversion (#10562)

ggml-ci

b4221

ggml-cpu: fix typo in gemv/gemm iq4_nl_4_4 (#10580)

b4220

sycl : offload of get_rows set to 0 (#10432)

b4219

sycl : Reroute permuted mul_mats through oneMKL (#10408)

This PR fixes the failing MUL_MAT tests for the sycl backend.

b4218

CANN: RoPE operator optimization (#10563)

* [cann] RoPE operator optimization

* [CANN]Code Formatting

---------

Co-authored-by: noemotiovon <[email protected]>