Load all MoE experts during warmup #11571
llama : use all experts during warmup
83a473a0
llama : increased max_nodes as large MoE models use massive amounts o…
c8bc6e4f
Merge remote-tracking branch 'upstream/master' into experts-warmup
21213352
llama : correct llama_set_warmup() description
334515f6
ggerganov
approved these changes
on 2025-03-14
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub