Skip to content

Optimize MOE GEMV kernel for BS > 1. #20905

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
JohannesGaessler merged 4 commits into from
Mar 29, 2026
Merged

Optimize MOE GEMV kernel for BS > 1. #20905

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c5ab1c6
Optimize MOE GEMV kernel for BS > 1.
gaugarg-nv Mar 23, 2026
e3c02be
Remove em-dashes
gaugarg-nv Mar 23, 2026
5d282ad
Cherry-pick changes from @am17an PR https://github.com/ggml-org/llama…
am17an Mar 23, 2026
db58c0a
Make the max batch size for MOE GEMV kernel configurable based on GPU…
gaugarg-nv Mar 29, 2026
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
19 changes: 12 additions & 7 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 @@ -2338,7 +2338,8 @@ static void ggml_cuda_mul_mat_id(ggml_backend_cuda_context & ctx, ggml_tensor *
static_assert(MMVQ_MAX_BATCH_SIZE == MMVF_MAX_BATCH_SIZE);
if (ne2 <= MMVQ_MAX_BATCH_SIZE) {
if (ggml_is_quantized(src0->type)) {
if (ne2 <= MMVQ_MMID_MAX_BATCH_SIZE) {
const int mmvq_mmid_max = get_mmvq_mmid_max_batch(src0->type, cc);
if (ne2 <= mmvq_mmid_max) {
ggml_cuda_mul_mat_vec_q(ctx, src0, src1, ids, dst);
return;
}
Expand Down Expand Up @@ -2941,14 +2942,18 @@ static bool ggml_cuda_graph_check_compability(ggml_cgraph * cgraph) {
}

// [TAG_MUL_MAT_ID_CUDA_GRAPHS]
if (node->op == GGML_OP_MUL_MAT_ID && (!ggml_is_quantized(node->src[0]->type) || node->ne[2] > MMVQ_MMID_MAX_BATCH_SIZE)) {
// under these conditions, the mul_mat_id operation will need to synchronize the stream, so we cannot use CUDA graphs
// TODO: figure out a way to enable for larger batch sizes, without hurting performance
// ref: https://github.com/ggml-org/llama.cpp/pull/18958
use_cuda_graph = false;
if (node->op == GGML_OP_MUL_MAT_ID) {
const int cc = ggml_cuda_info().devices[ggml_cuda_get_device()].cc;
const int mmvq_mmid_max = get_mmvq_mmid_max_batch(node->src[0]->type, cc);
if (!ggml_is_quantized(node->src[0]->type) || node->ne[2] > mmvq_mmid_max) {
// under these conditions, the mul_mat_id operation will need to synchronize the stream, so we cannot use CUDA graphs
// TODO: figure out a way to enable for larger batch sizes, without hurting performance
// ref: https://github.com/ggml-org/llama.cpp/pull/18958
use_cuda_graph = false;
#ifndef NDEBUG
GGML_LOG_DEBUG("%s: disabling CUDA graphs due to unsupported node type\n", __func__);
GGML_LOG_DEBUG("%s: disabling CUDA graphs due to unsupported node type\n", __func__);
#endif
}
}

if (!use_cuda_graph) {
Expand Down
Loading