llama.cpp
CUDA: mul_mat_id for mmf for bs <= 64 for f16 and bs <= 32 for f32
#16277

Merged

CUDA: mul_mat_id for mmf for bs <= 64 for f16 and bs <= 32 for f32 #16277

JohannesGaessler merged 2 commits into ggml-org:master from am17an:cuda_mmf_mmid

CUDA: mul_mat_id for mmf for bs <= 64 for f16 and bs <= 32 for f32

8ee101c0

am17an requested a review from

slaren 303 days ago

am17an requested a review from

JohannesGaessler 303 days ago

github-actions added testing

github-actions added Nvidia GPU

github-actions added ggml

JohannesGaessler commented on 2025-09-26

Review: refactor if statement

6aef41df

slaren approved these changes on 2025-09-27

JohannesGaessler approved these changes on 2025-09-27

JohannesGaessler merged c0bfc57a into master 301 days ago

am17an deleted the cuda_mmf_mmid branch 301 days ago

Reviewers

JohannesGaessler

slaren

Assignees

No one assigned

Labels

testing Nvidia GPU ggml

Milestone

No milestone