Merge pull request #3 from openradx/fix-llamacpp_server

Fix llama.cpp server by building it in CI

.github/workflows/ci.yml

@@ -19,7 +19,15 @@ jobs:
         uses: actions/checkout@v3
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v2
-      - name: Build and cache Docker images
+      - name: Build and cache llama.cpp server image
+        uses: docker/build-push-action@v3
+        with:
+          file: ./compose/llamacpp/llamacpp.Dockerfile
+          load: true
+          tags: llamacpp-server:latest
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+      - name: Build and cache main RADIS image
         uses: docker/build-push-action@v3
         with:
           context: .

compose/docker-compose.base.yml

@@ -53,11 +53,11 @@ services:
         celery --broker=amqp://rabbit/ flower --url_prefix=flower
       "
 
-  # llamacpp:
-  #   build:
-  #     context: llamacpp
-  #     dockerfile: llamacpp.Dockerfile
-  #   hostname: llamacpp.local
+  llamacpp:
+    build:
+      context: llamacpp
+      dockerfile: llamacpp.Dockerfile
+    hostname: llamacpp.local
 
   postgres:
     image: postgres:16.2

compose/docker-compose.dev.yml

@@ -64,17 +64,18 @@ services:
     profiles:
       - full
 
-  # llamacpp:
-  #   ports:
-  #     - 9610:8080
-  #   volumes:
-  #     - ../models:/models
-  #   command: >
-  #     bash -c '
-  #       /server -m /models/model.gguf -c 512 --host 0.0.0.0 --port 8080
-  #     '
-  #   profiles:
-  #     - full
+  llamacpp:
+    image: llamacpp-server:latest
+    ports:
+      - 9610:8080
+    volumes:
+      - ../models:/models
+    command: >
+      bash -c '
+        /server -m /models/model.gguf -c 512 --host 0.0.0.0 --port 8080
+      '
+    profiles:
+      - full
 
   postgres:
     environment:

compose/docker-compose.prod.yml

@@ -58,6 +58,7 @@ services:
       replicas: 1
 
   llamacpp:
+    image: llamacpp-server:latest
     restart: always
     volumes:
       - models_data:/models