ggml : persistent expert cache for --n-cpu-moe (RFC #20757) - #21614

Closed

e1n00r wants to merge 3 commits into

ggml-org:masterfrom

e1n00r:feat/expert-cache-v2

Commits on Apr 8, 2026

implement key structs and logic for expert cache
e1n00r
committed
f715d6d
add auto sizing and resolve --n-cpu-moe conflicts
e1n00r
committed
ff16f5a
add support for llama-bench
e1n00r
committed
0e5d061