llama.cpp
ggml-cpu: use LUT for converting e8->f32 scales on x86
#19288

Merged

ggml-cpu: use LUT for converting e8->f32 scales on x86 #19288

am17an merged 2 commits into ggml-org:master from am17an:mxfp4-cpu-scale

ggml-cpu: use LUT for converting e8->f32 scales on x86

635c8df4

am17an requested a review from

ggerganov 7 days ago

ggerganov commented on 2026-02-03

github-actions added ggml

add dispatch based on macro

e0a63352

ggerganov approved these changes on 2026-02-03

am17an merged 2ceda3f6 into master 6 days ago

am17an deleted the mxfp4-cpu-scale branch 6 days ago

Reviewers

ggerganov

Assignees

No one assigned

Labels

ggml

Milestone

No milestone