Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
10,849 Commits 5 Branches 0 Tags
e71b80510c848c00175924ecf3c40333ccae8eb5
Commit Graph
3 Commits
Author SHA1 Message Date
Piotr Wilkin (ilintar) e71b80510c Revert "CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#24546)" (#28551) 
This reverts commit 0c963452ea.

Assisted-by: Claude Fable 5.1
Claude-Session: https://claude.ai/code/session_01Q7rfnkjzgnfhvJsdeXhdoH
 2026-09-07 16:28:19 +02:00
ravel7524andCarl Philipp Klemm 0c963452ea CUDA: size routed MoE MMQ N-tiles from typical expert width on RDNA3 (#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>
 2026-09-07 15:22:42 +02:00
Johannes Gäßler 6eddde06a4 CUDA: refactor MMQ kernel configuration (#24127) 
* CUDA: refactor MMQ kernel configuration

* fix Blackwell config

* remove legacy code
 2026-07-13 18:37:57 +02:00