llama.cpp
metal : optimize sparse FA + clean-up
#29377
Merged

metal : optimize sparse FA + clean-up #29377

ggerganov merged 4 commits into master from gg/metal-fa-sparse-opt
ggerganov
ggerganov ggerganov requested a review 33 days ago
github-actions github-actions added documentation
github-actions github-actions added ggml
github-actions github-actions added Apple Metal
ggerganov metal : cache sparse FA indices in shared memory
a1a4255e
ggerganov metal : simplify shared memory size calculation
7c8c8f7a
ggerganov pi : update general
ecad2465
ggerganov metal : unroll sparse index load
4c9f0cbe
ggerganov ggerganov force pushed from faf2fb82 to 4c9f0cbe 33 days ago
ggerganov ggerganov merged cdc06426 into master 33 days ago
nikwen
ggerganov ggerganov deleted the gg/metal-fa-sparse-opt branch 30 days ago

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone