llama.cpp
ggml-cpu: aarm64: q4_K repack gemm and gemv implementations (dotprod only)
#17494
Merged

ggml-cpu: aarm64: q4_K repack gemm and gemv implementations (dotprod only) #17494

Alcpz
Alcpz Enabled q4_K_4x8 path
37a477ca
Alcpz Fixed generic Q4_K 8x4 implementation
968d2d02
Alcpz wip: dotprod gemm
4f0cee7d
Alcpz Working arm q4_K dotprod gemm
328512b6
Alcpz Undo acc rename
2ce46b8e
Alcpz Q4_K arm dotprod gemm
6e7caead
Alcpz Fix: q4_qs reinterpret from uint to int
66d66519
Alcpz Removed comments
bf717d90
Alcpz Fixed macro guards
1f7d3ebe
Alcpz Fixed unused vars in generic implementation
9aecf340
Alcpz Fixed unused vars in 8x4 repack
ccb84f6e
Alcpz Fixed unused vars in generic implementation, unneeded comment
dfbc4c65
Alcpz Alcpz requested a review from ggerganov ggerganov 36 days ago
github-actions github-actions added ggml
Alcpz Missing arch fallback for x86
eb94631c
Alcpz
ggerganov
Alcpz
ggerganov
ggerganov approved these changes on 2025-11-27
ggerganov minor : style
a2c991c0
ggerganov ggerganov merged cd8370b4 into master 34 days ago
Alcpz Alcpz deleted the Alcpz/arm_q4_k_repack_dotprod branch 34 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone