diff --git a/onnxruntime/core/providers/cuda/integer_gemm.cc b/onnxruntime/core/providers/cuda/integer_gemm.cc
index 76bc708736..9c2a05e9a9 100644
--- a/onnxruntime/core/providers/cuda/integer_gemm.cc
+++ b/onnxruntime/core/providers/cuda/integer_gemm.cc
@@ -20,8 +20,9 @@ Status GemmInt8(int m, int n, int k,
                 const CudaKernel* cuda_kernel, onnxruntime::Stream* ort_stream) {
   ORT_ENFORCE(a != nullptr && b != nullptr && c != nullptr, "input matrix should not be null");
   ORT_ENFORCE(cuda_kernel != nullptr, "kernel is null");
+  ORT_ENFORCE(ort_stream != nullptr, "Cuda kernel must have the stream instance");
 
-  cudaStream_t stream = ort_stream ? static_cast<cudaStream_t>(ort_stream->GetHandle()) : nullptr;
+  cudaStream_t stream = static_cast<cudaStream_t>(ort_stream->GetHandle());
 
   // pad A and B to make their leading dimension be multiples of 32
   // because cublasGemmEx requires: