fix(docker): Use localrecall main tag instead of master

Signed-off-by: Richard Palethorpe <io@richiejp.com>

docker-compose.yaml

@@ -5,16 +5,16 @@ services:
     # Available images with CUDA, ROCm, SYCL, Vulkan
     # Image list (quay.io): https://quay.io/repository/go-skynet/local-ai?tab=tags
     # Image list (dockerhub): https://hub.docker.com/r/localai/localai
-    image: localai/localai:latest-cpu
+    image: local-ai
     command: 
-    # - rombo-org_rombo-llm-v3.0-qwen-32b # minimum suggested model
+    - rombo-org_rombo-llm-v3.0-qwen-32b # minimum suggested model
     - arcee-agent # (smaller)
     - granite-embedding-107m-multilingual
     healthcheck:
       test: ["CMD", "curl", "-f", "http://localhost:8080/readyz"]
-      interval: 10s
-      timeout: 20m
-      retries: 20
+      interval: 60s
+      timeout: 10m
+      retries: 120
     ports:
     - 8081:8080
     environment:
@@ -23,6 +23,9 @@ services:
     volumes:
       - ./volumes/models:/build/models:cached
       - ./volumes/images:/tmp/generated/images
+    devices:
+      - /dev/dri
+
     # decomment the following piece if running with Nvidia GPUs
     # deploy:
     #   resources:
@@ -32,7 +35,7 @@ services:
     #           count: 1
     #           capabilities: [gpu]
   localrecall:
-    image: quay.io/mudler/localrecall:master
+    image: quay.io/mudler/localrecall:main
     ports:
       - 8080
     environment: