diff --git a/README.md b/README.md
index a880023..b6d55ab 100644
--- a/README.md
+++ b/README.md
@@ -12,7 +12,7 @@ Instead of starting torchserve as given in the example, start it with a docker c
 
 ```docker run -v $(pwd)/model_store:/model_store -p 8080:8080 --rm --name torchserve -d iqtlabs/torchserve --models <model>=<model>.mar```
 
-* [iqtlabs/cuda-torchserve](https://hub.docker.com/r/iqtlabs/cuda-torchserve): CUDA (11.8 or later) accelerated for amd64 only.
+* [iqtlabs/cuda-torchserve](https://hub.docker.com/r/iqtlabs/cuda-torchserve): CUDA (12.1 or later) accelerated for amd64 only.
 
 ```docker run --gpus all -v $(pwd)/model_store:/model_store -p 8080:8080 --rm --name torchserve -d iqtlabs/cuda-torchserve --models <model>=<model>.mar```
 
diff --git a/docker/Dockerfile.orin-torchserve b/docker/Dockerfile.orin-torchserve
index 1c65ae2..10e22e3 100644
--- a/docker/Dockerfile.orin-torchserve
+++ b/docker/Dockerfile.orin-torchserve
@@ -1,3 +1,4 @@
+# https://catalog.ngc.nvidia.com/orgs/nvidia/containers/l4t-pytorch
 FROM nvcr.io/nvidia/l4t-pytorch:r35.2.1-pth2.0-py3
 ARG PIP_CACHE
 ENV DEBIAN_FRONTEND=noninteractive