diff --git a/compose.yml b/compose.yml
index 459bbcb..f5ab06e 100644
--- a/compose.yml
+++ b/compose.yml
@@ -87,14 +87,14 @@ services:
 
   app:
     user: "${DOCKER_USER:-1000}"
-    image: lasuite/drive-frontend:v0.11.1
+    image: lasuite/drive-frontend:v0.12.0
     networks:
       - backend
     deploy:
       labels:
         - "traefik.enable=false"
         - "coop-cloud.${STACK_NAME}.timeout=${TIMEOUT:-120}"
-        - "coop-cloud.${STACK_NAME}.version=0.3.0+v0.11.1"
+        - "coop-cloud.${STACK_NAME}.version=0.4.0+v0.12.0"
     environment:
       <<: [ *common-env ]
     healthcheck:
@@ -108,7 +108,7 @@ services:
 
   backend:
     user: ${DOCKER_USER:-1000}
-    image: lasuite/drive-backend:v0.11.1
+    image: lasuite/drive-backend:v0.12.0
     command: [ "gunicorn", "-c", "/usr/local/etc/gunicorn/drive.py", "drive.wsgi:application" ]
     entrypoint: [ "/abra-entrypoint.sh", "/usr/local/bin/entrypoint" ]
     environment:
@@ -133,7 +133,7 @@ services:
   # this may be unecessary once the PR is merged to use celery beat
   # but this is a workaround for now https://github.com/suitenumerique/drive/issues/484
   scheduler:
-    image: lasuite/drive-backend:v0.11.1
+    image: lasuite/drive-backend:v0.12.0
     user: ${DOCKER_USER:-1000}
     networks:
       - backend
@@ -160,7 +160,7 @@ services:
 
   celery:
     user: ${DOCKER_USER:-1000}
-    image: lasuite/drive-backend:v0.11.1
+    image: lasuite/drive-backend:v0.12.0
     networks:
       - backend
     command: [ "celery", "-A", "drive.celery_app", "worker", "-l", "INFO" ]
@@ -207,7 +207,7 @@ services:
       - postgres_p
 
   redis:
-    image: redis:5
+    image: redis:8
     networks:
       - backend 
 
@@ -350,7 +350,7 @@ services:
 
 
   web:
-    image: nginx:1.25
+    image: nginx:1.29
     configs:
       - source: nginx_conf
         target: /etc/nginx/conf.d/default.conf