diff --git a/minimal-ml-inference/README.md b/minimal-ml-inference/README.md
index 988679b..7b22e90 100755
--- a/minimal-ml-inference/README.md
+++ b/minimal-ml-inference/README.md
@@ -107,10 +107,10 @@ docker save $MODEL_NAME | docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_POR
 With the application image on the device, it can be started using `docker-compose.yml`:
 
 ```sh
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose up
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP up
 
 # Terminate with Ctrl-C and cleanup
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose down --volumes
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP down --volumes
 ```
 
 The expected output from the application is the raw predictions from the model specified in the environment variable.
diff --git a/object-detector-cpp/README.md b/object-detector-cpp/README.md
index a9c0644..6042e0e 100644
--- a/object-detector-cpp/README.md
+++ b/object-detector-cpp/README.md
@@ -116,10 +116,10 @@ docker save $MODEL_NAME | docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_POR
 With the images on the device, they can be started using `docker-compose.yml`:
 
 ```sh
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose up
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP up
 
 # Terminate with Ctrl-C and cleanup
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose down --volumes
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP down --volumes
 ```
 
 ### The expected output
diff --git a/object-detector-python/README.md b/object-detector-python/README.md
index 760793e..23f072a 100755
--- a/object-detector-python/README.md
+++ b/object-detector-python/README.md
@@ -129,10 +129,10 @@ docker save $MODEL_NAME | docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_POR
 With the application image on the device, it can be started using `docker-compose.yml`:
 
 ```sh
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose up
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP up
 
 # Terminate with Ctrl-C and cleanup
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose down --volumes
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP down --volumes
 ```
 
 ### The expected output
diff --git a/pose-estimator-with-flask/README.md b/pose-estimator-with-flask/README.md
index 11a1732..bffe290 100755
--- a/pose-estimator-with-flask/README.md
+++ b/pose-estimator-with-flask/README.md
@@ -142,10 +142,10 @@ docker save $MODEL_NAME | docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_POR
 With the application image on the device, it can be started using `docker-compose.yml`:
 
 ```sh
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose up
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP up
 
 # Terminate with Ctrl-C and cleanup
-docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT --env-file ./config/env.$ARCH.$CHIP compose down --volumes
+docker --tlsverify --host tcp://$DEVICE_IP:$DOCKER_PORT compose --env-file ./config/env.$ARCH.$CHIP down --volumes
 ```
 
 ### The expected output