b3799

ggml-alloc : fix list of allocated tensors with GGML_ALLOCATOR_DEBUG …

b3798

Update CUDA graph on scale change plus clear nodes/params  (#9550)

* Avoid using saved CUDA graph if scale changes and reset nodes/params on update

Fixes https://github.com/ggerganov/llama.cpp/issues/9451

* clear before resize

b3796

quantize : improve type name parsing (#9570)

quantize : do not ignore invalid types in arg parsing

quantize : ignore case of type and ftype arguments

b3795

ggml : fix builds (#0)

ggml-ci

b3790

CUDA: fix sum.cu compilation for CUDA < 11.7 (#9562)

b3787

server : clean-up completed tasks from waiting list (#9531)

ggml-ci

b3785

ggml : fix n_threads_cur initialization with one thread (#9538)

* ggml : fix n_threads_cur initialization with one thread

* Update ggml/src/ggml.c

---------

Co-authored-by: Max Krasnyansky <[email protected]>

b3772

ggml : move common CPU backend impl to new header (#9509)

b3749

feat: remove a sampler from a chain (#9445)

* feat: remove a sampler from a chain

* fix: return removed sampler

* fix: safer casting

b3733

llama : skip token bounds check when evaluating embeddings (#9437)