diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index 598f3228c..56eeb757a 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -4021,7 +4021,7 @@ static void ggml_cuda_graph_evaluate_and_capture(ggml_backend_cuda_context * cud
 
             for (int i = 1; i <= concurrent_event->n_streams; ++i) {
                 cudaStream_t stream = cuda_ctx->stream(cuda_ctx->device, i);
-                CUDA_CHECK(cudaStreamWaitEvent(stream, concurrent_event->fork_event));
+                CUDA_CHECK(cudaStreamWaitEvent(stream, concurrent_event->fork_event, 0));
             }
         }
     };
@@ -4105,7 +4105,7 @@ static void ggml_cuda_graph_evaluate_and_capture(ggml_backend_cuda_context * cud
                             // Wait on join events of forked streams in the main stream
                             CUDA_CHECK(cudaEventRecord(concurrent_event->join_events[i - 1],
                                                        cuda_ctx->stream(cuda_ctx->device, i)));
-                            CUDA_CHECK(cudaStreamWaitEvent(cuda_ctx->stream(), concurrent_event->join_events[i - 1]));
+                            CUDA_CHECK(cudaStreamWaitEvent(cuda_ctx->stream(), concurrent_event->join_events[i - 1], 0));
                         }
 
                         is_concurrent_event_active = false;