llama.cpp
llama : quantize up to 31% faster on Linux with mmap
#3206

Merged

llama : quantize up to 31% faster on Linux with mmap #3206

ggerganov merged 3 commits into ggml-org:master from cebtenzzre:faster-quantize-2

llama : enable mmap in quantize on Linux -> 31% faster

32bc3f4f

cebtenzzre requested a review from

ggerganov 2 years ago

cebtenzzre requested a review from

ikawrakow 2 years ago

slaren commented on 2023-09-23

slaren approved these changes on 2023-09-23

also enable mmap on Windows

b3a6b286

Merge branch 'master' into HEAD

66382f1e

ggerganov approved these changes on 2023-09-29

ggerganov merged 2777a84b into master 2 years ago

Reviewers

ggerganov

slaren

ikawrakow

Assignees

No one assigned

Labels

None yet

Milestone

No milestone