diff --git a/examples/onnx/yolov5/README.md b/examples/onnx/yolov5/README.md
new file mode 100644
index 0000000..cb664c6
--- /dev/null
+++ b/examples/onnx/yolov5/README.md
@@ -0,0 +1,15 @@
+# YOLOV5 DEMO
+
+### 1.About onnx file
+
+- This model remove the post-process layer(which is not available on NPU now).
+- To export .onnx format model file, refer to https://github.com/airockchip/rknn_model_zoo/tree/main/models/vision/object_detection/yolov5-pytorch
+
+
+
+### 2.Run demo
+
+```
+python test.py
+```
+
diff --git a/examples/onnx/yolov5/test.py b/examples/onnx/yolov5/test.py
index d8cc442..f9b4b05 100644
--- a/examples/onnx/yolov5/test.py
+++ b/examples/onnx/yolov5/test.py
@@ -236,7 +236,7 @@ if __name__ == '__main__':
 
     # Load ONNX model
     print('--> Loading model')
-    ret = rknn.load_onnx(model=ONNX_MODEL, outputs=['378', '439', '500'])
+    ret = rknn.load_onnx(model=ONNX_MODEL)
     if ret != 0:
         print('Load model failed!')
         exit(ret)
@@ -301,8 +301,8 @@ if __name__ == '__main__':
     if boxes is not None:
         draw(img_1, boxes, scores, classes)
     # show output
-    # cv2.imshow("post process result", img_1)
-    # cv2.waitKey(0)
-    # cv2.destroyAllWindows()
+    cv2.imshow("post process result", img_1)
+    cv2.waitKey(0)
+    cv2.destroyAllWindows()
 
     rknn.release()
diff --git a/examples/onnx/yolov5/yolov5s.onnx b/examples/onnx/yolov5/yolov5s.onnx
index 094f5f1..6165589 100644
Binary files a/examples/onnx/yolov5/yolov5s.onnx and b/examples/onnx/yolov5/yolov5s.onnx differ