b5361

llama-bench : add defrag-thold, check for invalid ranges (#13487)

b5360

opencl: remove unnecessary assert for `add` (#13257)

b5359

clip : cap max image size 1024 for qwen vl model (#13478)

b5358

llama/ggml: add LLM training support (#10544)

* llama/ggml: add LLM training support

more compact progress bar

llama_save_model_to_file

llama_opt_param_filter

ggml_graph_dup force_grads

refactor ggml_opt, fix test-opt

* remove logits_all

* refactor CUDA implementation for ACC

* reset graph at beginning of opt period

b5357

context : fix state io for memory-less contexts (#13470)

ggml-ci

b5356

server : allow content to be null in oaicompat_completion_params_pars…

b5355

llama-bench : accept ranges for integer parameters (#13410)

b5354

ggml-cpu: Integrate fp32=bf16xbf16 SME KleidiAI kernel (#13053)

* ggml-cpu: Integrate fp32=bf16xbf16 SME KleidiAI kernel

Signed-off-by: Dan Johansson <dan.johansson@arm.com>

* * code review fixes

Signed-off-by: Dan Johansson <dan.johansson@arm.com>

* * adds a comment that clarifies barrier usage

Signed-off-by: Dan Johansson <dan.johansson@arm.com>

---------

Signed-off-by: Dan Johansson <dan.johansson@arm.com>
Co-authored-by: Charles Xu <charles.xu@arm.com>

b5353

CUDA: fix misaligned synchronization in FA (#13469)

b5352

ggml : add mrope kernel for metal (#13457)