diff --git a/paddle/fluid/inference/api/analysis_predictor.cc b/paddle/fluid/inference/api/analysis_predictor.cc
index 7badcb395ea708b621b7aa62289d9ff830550858..7ec3271c66573ebf91b663aed9420295a8a8bdb0 100644
--- a/paddle/fluid/inference/api/analysis_predictor.cc
+++ b/paddle/fluid/inference/api/analysis_predictor.cc
@@ -1077,6 +1077,12 @@ std::unique_ptr<PaddlePredictor> CreatePaddlePredictor<
         process_level_allocator_enabled = true;
       }
 
+      // TODO(Jingzhuangzhuang): Fix trt error when allocator_strategy is
+      // auto_growth
+      if (config.tensorrt_engine_enabled()) {
+        gflags.push_back("--allocator_strategy=naive_best_fit");
+      }
+
       if (framework::InitGflags(gflags)) {
         VLOG(3) << "The following gpu analysis configurations only take effect "
                    "for the first predictor: ";