diff --git a/ggml/src/ggml-cuda/ggml-cuda.cu b/ggml/src/ggml-cuda/ggml-cuda.cu
index 67b5e16d8..0f88dfe88 100644
--- a/ggml/src/ggml-cuda/ggml-cuda.cu
+++ b/ggml/src/ggml-cuda/ggml-cuda.cu
@@ -762,7 +762,8 @@ static enum ggml_status ggml_backend_cuda_buffer_init_tensor(ggml_backend_buffer
 
         if (padded_size > original_size) {
             ggml_cuda_set_device(ctx->device);
-            CUDA_CHECK(cudaMemset((char *)tensor->data + original_size, 0, padded_size - original_size));
+            CUDA_CHECK(cudaMemsetAsync((char *)tensor->data + original_size, 0, padded_size - original_size, cudaStreamPerThread));
+            CUDA_CHECK(cudaStreamSynchronize(cudaStreamPerThread));
         }
     }
     return GGML_STATUS_SUCCESS;