b5747

CUDA/HIP: optimize mmv paths taken for HIP devices (#14324)

Co-authored-by: Johannes Gäßler <[email protected]>

b5745

vulkan: update windows SDK in release.yml (#14344)

b5744

llama : better rwkv chat template and add missing `inputs.use_jinja` …

b5743

CUDA: mul_mat_v support for batch sizes > 1 (#14262)

* CUDA: mul_mat_v support for batch sizes > 1

* use 64 bit math for initial offset calculation

b5742

kv-cells : fix tracking of seq_pos (#14339)

* kv-cells : fix tracking of seq_pos during cache reuse

ggml-ci

* cont : improve error message

ggml-ci

* cont : add more comments

b5740

quantize : handle user-defined pruning of whole layers (blocks) (#13037)

b5738

run : avoid double tokenization (#14327)

* run : avoid double tokenization by adopting common_tokenize heuristic

* build : fix windows gcc and clang warnings

* lint : fixed trailing whitepace

* run : fix is_first flag

b5737

examples : fix is_first logic for tokenization (#14329)

ggml-ci

b5736

HIP: enable vec fattn on RDNA4 (#14323)

b5735

mtmd : fix Pixtral OOM with large images by capping image_size to 102…