ravel7524andCarl Philipp Klemm 4c5a9a02a1 CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (llama/24546) ...

* adjust ncols_picker for routed MoE in mul_mat_q_case function

* Adding CDNA, RDNA2 and RDNA4

* fix: update mmq_use_routed_moe_ncols_picker to include NVIDIA + Volta support

* feat: enhance mmq configuration for various architectures with moe_ncols_min_cc support

* refactor: replace moe_ncols_min_cc with use_typical_moe_ncols in mmq configuration files

* HIP: mmq: enable typical moe ncols on RDNA4

---------

Co-authored-by: Carl Philipp Klemm <carl@uvos.xyz>

2026-09-14 20:45:06 +03:00

..

cmake

kleidiai: Rework KleidiAI Build System/Integration (llama/26077)

2026-09-04 13:39:29 +03:00

include

ggml : rename and make private ggml_op_alloc_size_may_expand() (ggml/0)

2026-09-04 13:39:43 +03:00

src

CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (llama/24546)

2026-09-14 20:45:06 +03:00

.gitignore

…

CMakeLists.txt

ggml : bump version to 0.23.0 (ggml/1618)

2026-09-04 13:39:43 +03:00