diff --git a/README.md b/README.md
index 9964753..f8950ee 100644
--- a/README.md
+++ b/README.md
@@ -19,10 +19,11 @@ All the models are implemented in pytorch or mxnet first, and export a weights f
 ## Tutorials
 
 - [A guide for quickly getting started, taking lenet5 as a demo.](./tutorials/getting_started.md)
+- [Frequently Asked Questions (FAQ)](./tutorials/faq.md)
 - [Migrating from TensorRT 4 to 7](./tutorials/migrating_from_tensorrt_4_to_7.md)
 - [How to implement multi-GPU processing, taking YOLOv4 as example](./tutorials/multi_GPU_processing.md)
 - [Check if Your GPU support FP16/INT8](./tutorials/check_fp16_int8_support.md)
-- [Frequently Asked Questions (FAQ)](./tutorials/faq.md)
+- [Deploy YOLOv4 with Triton Inference Server](https://github.com/isarsoft/yolov4-triton-tensorrt)
 
 ## Test Environment