diff --git a/onnxruntime/core/providers/cuda/cuda_execution_provider.cc b/onnxruntime/core/providers/cuda/cuda_execution_provider.cc
index 514d6d50d6..d4e5d39b15 100644
--- a/onnxruntime/core/providers/cuda/cuda_execution_provider.cc
+++ b/onnxruntime/core/providers/cuda/cuda_execution_provider.cc
@@ -198,6 +198,8 @@ void CUDAExecutionProvider::AddDeferredReleaseCPUPtr(void* p) {
 }
 
 Status CUDAExecutionProvider::OnRunStart() {
+  // always set CUDA device when session::Run() in case it runs in a worker thread
+  CUDA_RETURN_IF_ERROR(cudaSetDevice(GetDeviceId()));
   auto cpu_alloc = GetAllocator(0, OrtMemTypeCPU);
   // check if cudaEvents has passed for deferred release
   // note that we need to take a mutex in case of multi-threaded Run()