Bedrock

.github/workflows/ci-chat-server.yml

@@ -22,6 +22,7 @@ jobs:
         uses: actions/checkout@v5
 
       - name: Ollama cache
+        if: false # ollama disabled in this branch
         uses: actions/cache@v4
         with:
           path: ./chat-server/.ollama
@@ -42,4 +43,5 @@ jobs:
         run: ./gradlew build --no-daemon --info
 
       - name: Ollama cache permissions
+        if: false # ollama disabled in this branch
         run: sudo chown -R $USER:$USER ./chat-server/.ollama

chat-server/build.gradle.kts

@@ -32,11 +32,11 @@ dependencies {
     implementation("org.springframework.ai:spring-ai-advisors-vector-store")
 
     // ollama
-    implementation("org.springframework.ai:spring-ai-starter-model-ollama")
+    // implementation("org.springframework.ai:spring-ai-starter-model-ollama")
 
     // bedrock
-    // implementation("org.springframework.ai:spring-ai-starter-model-bedrock")
-    // implementation("org.springframework.ai:spring-ai-starter-model-bedrock-converse")
+    implementation("org.springframework.ai:spring-ai-starter-model-bedrock")
+    implementation("org.springframework.ai:spring-ai-starter-model-bedrock-converse")
 
     implementation("org.springframework.ai:spring-ai-starter-vector-store-pgvector")
     runtimeOnly("org.postgresql:postgresql")

chat-server/docker-compose.yml

@@ -10,6 +10,8 @@ services:
 
   ollama:
     image: ollama/ollama:0.12.3
+    profiles:
+      - disabled # ollama disabled in this branch
     volumes:
       - ./.ollama:/root/.ollama
     ports:

chat-server/src/main/resources/application-bedrock.yml

@@ -0,0 +1,21 @@
+spring:
+  ai:
+    model:
+      embedding: "bedrock-cohere"
+      chat: "bedrock-converse"
+    bedrock:
+      aws:
+        access-key: "${AWS_ACCESS_KEY_ID}"
+        secret-key: "${AWS_SECRET_ACCESS_KEY}"
+        region: "${AWS_REGION:eu-central-1}"
+      converse:
+        chat:
+          options:
+            model: "${AWS_BEDROCK_CHAT_MODEL}"
+      cohere:
+        embedding:
+          model: "${AWS_BEDROCK_EMBEDDING_MODEL}"
+    vectorstore:
+      pgvector:
+        dimensions: 1024
+        initialize-schema: true

chat-server/src/main/resources/application.yml

@@ -1,6 +1,6 @@
 spring:
   profiles:
-    active: "ollama"
+    active: "bedrock"
   application:
     name: chat-server
   datasource:

chat-server/src/test/kotlin/com/rogervinas/ChatServerApplicationTest.kt

@@ -36,9 +36,10 @@ import java.util.UUID
 
 
 @SpringBootTest(webEnvironment = RANDOM_PORT)
-@ActiveProfiles("test", "ollama")
+@ActiveProfiles("test", "bedrock")
 @TestMethodOrder(MethodOrderer.OrderAnnotation::class)
 @Testcontainers
+@DisabledIfCI
 class ChatServerApplicationTest {
 
     companion object {
@@ -47,7 +48,7 @@ class ChatServerApplicationTest {
         val container = ComposeContainer(File("docker-compose.yml"))
             .withLocalCompose(true)
             .withExposedService("vectordb", 5432, forLogMessage(".*database system is ready to accept connections.*", 1))
-            .withExposedService("ollama", 11434, forLogMessage(".*inference compute.*", 1))
+            // .withExposedService("ollama", 11434, forLogMessage(".*inference compute.*", 1))
     }
 
     @Autowired