b5380

server : passthrough the /models endpoint during loading (#13535)

* server : passthrough the /models endpoint during loading

* server : update readme + return json for "meta" field

b5379

server : fix cache_tokens bug with no cache_prompt (#13533)

b5378

cmake: simplify vulkan shader test logic (#13263)

b5377

vulkan: KHR_coopmat flash attention (#13506)

This shader uses coopmat1 to do the Q*K^T multiply. The P*V multiply is more
difficult for various reasons so I haven't done it. Performance for this
shader is around 2.5x better than for the scalar shader when doing prompt
processing. Some of the benefit may be from other optimizations like staging
through shared memory, or splitting by rows.

b5372

vulkan: workaround FA compile failures on macos (#13517)

b5371

quantize : improve tensor-type pattern matching (#13033)

b5370

clip : clip.h become private API (⚠️ breaking change) (#13510)

b5369

metal : use FA-vec kernel up to batch size 20 (#13496)

* batched-bench : fix pp batch contents

* metal : optimize multi-sequence FA vec kernel

ggml-ci

* metal : use FA-vec kernel up to batch size 20

ggml-ci

b5368

metal : optimize multi-sequence FA vec kernel (#13493)

* batched-bench : fix pp batch contents

* metal : optimize multi-sequence FA vec kernel

ggml-ci

b5367

ggml-cpu: Update KleidiAI to v1.6 and fix include directives (#13509)

Signed-off-by: Dan Johansson <[email protected]>