Skip to content

cuda: fix rope fusion for gemma3 #17378

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 19, 2025
Merged

cuda: fix rope fusion for gemma3 #17378

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
cuda: fix rope fusion for gemma3
  • Loading branch information
@am17an
am17an committed Nov 19, 2025
commit 01590ff9563873765bfa396d85b0c70321595582
4 changes: 4 additions & 0 deletions ggml/src/ggml-cuda/ggml-cuda.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3001,6 +3001,10 @@ static void update_cuda_graph_executable(ggml_backend_cuda_context * cuda_ctx) {
static bool ggml_cuda_should_fuse_rope_set_rows(const ggml_tensor * rope,
const ggml_tensor * view,
const ggml_tensor * set_rows) {

if (rope->op != GGML_OP_ROPE || view->op != GGML_OP_VIEW || set_rows->op != GGML_OP_SET_ROWS) {
return false;
}
// ne3 not tested
if (rope->src[0]->ne[3] != 1) {
return false;
Expand Down
Loading