llama.cpp
CUDA: add set rows for f32 and f16
#14551
Merged

CUDA: add set rows for f32 and f16 #14551

ggerganov merged 4 commits into ggml-org:master from am17an:cuda_set_rows
am17an
github-actions github-actions added Nvidia GPU
github-actions github-actions added ggml
am17an am17an requested a review from JohannesGaessler JohannesGaessler 271 days ago
am17an am17an force pushed 271 days ago
ggerganov
ggerganov
ggerganov commented on 2025-07-07
am17an
JohannesGaessler
JohannesGaessler commented on 2025-07-07
ggerganov
slaren
JohannesGaessler
am17an
am17an am17an force pushed 269 days ago
JohannesGaessler
JohannesGaessler commented on 2025-07-08
github-actions github-actions added examples
am17an am17an force pushed 269 days ago
ggerganov
am17an CUDA: add set rows for f32 and f16
853bc5ec
am17an Review: change kernel params, use strides from host
15e1b897
am17an Use 1-d kernel
85e2a202
am17an am17an force pushed to 85e2a202 266 days ago
am17an
am17an am17an requested a review from JohannesGaessler JohannesGaessler 266 days ago
ggerganov
JohannesGaessler
JohannesGaessler commented on 2025-07-12
JohannesGaessler
am17an Review: use int64_t for blockDim.x, rename nb->s for clarity
9deb7644
am17an am17an requested a review from JohannesGaessler JohannesGaessler 266 days ago
JohannesGaessler
JohannesGaessler approved these changes on 2025-07-12
ggerganov ggerganov merged 7de5c7ca into master 265 days ago
am17an am17an deleted the cuda_set_rows branch 164 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone