llama.cpp
ggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le) hardware – 3-4x performance boost
#15385
Merged
Overview
Commits
3
Changes
View On GitHub

ggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le) hardware – 3-4x performance boost #15385

ggerganov merged 3 commits into ggml-org:master from mgiessing:mxfp4_ppc64le
mgiessing
mgiessing Added VSX intrinsics for Power9+ systems
b777851a
mgiessing Manual unrolling for minor perf improvement
537b5040
github-actions github-actions added ggml
mgiessing mgiessing changed the title Add VSX intrinsics for Power9+ (ppc64le) hardware – 4-5x performance boost Add VSX intrinsics for Power9+ (ppc64le) hardware – 3-4x performance boost 135 days ago
mgiessing mgiessing changed the title Add VSX intrinsics for Power9+ (ppc64le) hardware – 3-4x performance boost ggml-cpu: add mxfp4 VSX intrinsics for Power9+ (ppc64le) hardware – 3-4x performance boost 135 days ago
ggerganov
ggerganov approved these changes on 2025-08-18
mgiessing Update ggml/src/ggml-cpu/arch/powerpc/quants.c
a06338ea
ggerganov
ggerganov commented on 2025-08-19
ggerganov ggerganov merged 6424594c into master 134 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
ggml
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum