llama.cpp
Clarify default MMQ for CUDA and LLAMA_CUDA_FORCE_MMQ flag
#8115

Merged

Clarify default MMQ for CUDA and LLAMA_CUDA_FORCE_MMQ flag #8115

JohannesGaessler merged 2 commits into ggml-org:master from mmq-readme-update

Add message about int8 support

b6cd699d

mofosyne added Review Complexity : Low

slaren approved these changes on 2024-06-25

slaren requested a review from

JohannesGaessler 1 year ago

JohannesGaessler requested changes on 2024-06-25

Add suggestions from review

37ff7098

JohannesGaessler approved these changes on 2024-06-26

JohannesGaessler merged 88540445 into master 1 year ago

Reviewers

JohannesGaessler

slaren

Assignees

No one assigned

Labels

Review Complexity : Low

Milestone

No milestone