infrastructure
Llama adds CPU-offload option for dense FFN weights
The --n-cpu-ffn option is added to offload dense FFN weights of the first N layers on CPU.
1 sources · 1 verified claimsAI knowledge entity
B10645 Llama Add N
One evolving timeline of verified developments, without duplicating the underlying truth across pages.
1 verified events1 independent sourcesUpdated 27 Aug 2026
Verified timeline