Upgrade dependencies, fix imports, and fix MultiChat configuration

Author: pacphi

build.gradle

@@ -1,8 +1,8 @@
 plugins {
 	id 'java'
 	id 'org.cyclonedx.bom' version '2.2.0'
-	id 'com.gorylenko.gradle-git-properties' version '2.4.2'
-	id 'org.springframework.boot' version '3.4.3'
+	id 'com.gorylenko.gradle-git-properties' version '2.5.0'
+	id 'org.springframework.boot' version '3.4.4'
 	id 'io.spring.dependency-management' version '1.1.7'
 	id 'org.springdoc.openapi-gradle-plugin' version '1.9.0'
 }
@@ -31,10 +31,10 @@ dependencies {
 	implementation('org.springframework.boot:spring-boot-starter-actuator')
 	implementation('org.springframework.boot:spring-boot-starter-validation')
 	implementation 'org.springframework.boot:spring-boot-starter-web'
-	implementation 'org.springframework.ai:spring-ai-openai-spring-boot-starter'
+	implementation 'org.springframework.ai:spring-ai-starter-model-openai'
 	implementation 'io.github.springboot-addons:spring-boot-starter-httpclient5-actuator:1.1.1'
 	implementation 'io.github.springboot-addons:spring-boot-starter-httpclient5-resilience4j:1.1.1'
-	implementation 'org.springdoc:springdoc-openapi-starter-webmvc-ui:2.8.5'
+	implementation 'org.springdoc:springdoc-openapi-starter-webmvc-ui:2.8.6'
 	testImplementation 'org.springframework.boot:spring-boot-starter-test'
 	testImplementation 'org.assertj:assertj-core:3.27.3'
 	testRuntimeOnly 'org.junit.platform:junit-platform-launcher'

docs/ENDPOINTS.md

@@ -44,7 +44,7 @@ Transfer-Encoding: chunked
     {
         "content": "Sure! Here's a random joke for you:\n\nWhy don't skeletons fight each other?  \nBecause they don’t have the guts! 😄",
         "errorMessage": null,
-        "generationTokens": 29,
+        "completionTokens": 29,
         "modelName": "openai/gpt-4o-2024-11-20",
         "promptTokens": 12,
         "responseTime": "1s893ms",
@@ -54,7 +54,7 @@ Transfer-Encoding: chunked
     {
         "content": "What do you call a fish wearing a bowtie? So-fish-ticated!",
         "errorMessage": null,
-        "generationTokens": 0,
+        "completionTokens": 0,
         "modelName": "anthracite-org/magnum-v2-72b",
         "promptTokens": 0,
         "responseTime": "1s49ms",
@@ -64,7 +64,7 @@ Transfer-Encoding: chunked
     {
         "content": "Why don't scientists trust atoms? Because they make up everything.",
         "errorMessage": null,
-        "generationTokens": 14,
+        "completionTokens": 14,
         "modelName": "qwen/qvq-72b-preview",
         "promptTokens": 41,
         "responseTime": "913ms",
@@ -74,7 +74,7 @@ Transfer-Encoding: chunked
     {
         "content": "Why don't eggs tell jokes? They'd crack each other up!",
         "errorMessage": null,
-        "generationTokens": 15,
+        "completionTokens": 15,
         "modelName": "x-ai/grok-2-1212",
         "promptTokens": 11,
         "responseTime": "666ms",
@@ -84,7 +84,7 @@ Transfer-Encoding: chunked
     {
         "content": "Here's a random joke for you:\n\nWhy don't scientists trust atoms?\n\nBecause they make up everything!",
         "errorMessage": null,
-        "generationTokens": 27,
+        "completionTokens": 27,
         "modelName": "anthropic/claude-3.5-haiku-20241022",
         "promptTokens": 12,
         "responseTime": "1s390ms",
@@ -94,7 +94,7 @@ Transfer-Encoding: chunked
     {
         "content": "Here's a random joke for you:\n\n**Why don't sharks eat clowns?**\n\nBecause they taste funny![1]",
         "errorMessage": null,
-        "generationTokens": 25,
+        "completionTokens": 25,
         "modelName": "perplexity/llama-3.1-sonar-huge-128k-online",
         "promptTokens": 5,
         "responseTime": "3s108ms",
@@ -104,7 +104,7 @@ Transfer-Encoding: chunked
     {
         "content": "Here's one:\n\nWhat do you call a fake noodle?\n\nAn impasta!",
         "errorMessage": null,
-        "generationTokens": 17,
+        "completionTokens": 17,
         "modelName": "meta-llama/llama-3.3-70b-instruct",
         "promptTokens": 15,
         "responseTime": "1s703ms",
@@ -114,7 +114,7 @@ Transfer-Encoding: chunked
     {
         "content": "What do you call fake spaghetti? An impasta",
         "errorMessage": null,
-        "generationTokens": 13,
+        "completionTokens": 13,
         "modelName": "mistralai/mistral-large-2411",
         "promptTokens": 8,
         "responseTime": "725ms",
@@ -124,7 +124,7 @@ Transfer-Encoding: chunked
     {
         "content": " Why did the chicken cross the playground? To get to the other slide!",
         "errorMessage": null,
-        "generationTokens": 15,
+        "completionTokens": 15,
         "modelName": "pygmalionai/mythalion-13b",
         "promptTokens": 30,
         "responseTime": "5s288ms",
@@ -134,7 +134,7 @@ Transfer-Encoding: chunked
     {
         "content": "Sure! Here’s a random joke for you:\n\nWhy don’t skeletons fight each other?  \nBecause they don’t have the guts! 💀😂",
         "errorMessage": null,
-        "generationTokens": 34,
+        "completionTokens": 34,
         "modelName": "deepseek/deepseek-chat",
         "promptTokens": 8,
         "responseTime": "4s477ms",
@@ -144,7 +144,7 @@ Transfer-Encoding: chunked
     {
         "content": "Okay, here's a random one:\n\nWhy don't scientists trust atoms? \n\nBecause they make up everything!\n",
         "errorMessage": null,
-        "generationTokens": 27,
+        "completionTokens": 27,
         "modelName": "google/gemini-2.0-flash-exp:free",
         "promptTokens": 6,
         "responseTime": "1s94ms",
@@ -154,7 +154,7 @@ Transfer-Encoding: chunked
     {
         "content": "Of course! Here's a light-hearted joke for you:\n\nWhy did the scarecrow win an award?\n\nBecause he was outstanding in his field! \n\nHope that brought a smile to your face!",
         "errorMessage": null,
-        "generationTokens": 41,
+        "completionTokens": 41,
         "modelName": "amazon/nova-pro-v1",
         "promptTokens": 5,
         "responseTime": "1s83ms",

src/main/java/me/pacphi/config/Chat.java

@@ -0,0 +1,24 @@
+package me.pacphi.config;
+
+import org.springframework.boot.web.client.RestClientCustomizer;
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+import org.springframework.http.HttpHeaders;
+import org.springframework.http.client.SimpleClientHttpRequestFactory;
+
+import java.time.Duration;
+
+@Configuration
+public class Http {
+
+    @Bean
+    public RestClientCustomizer restClientCustomizer() {
+        return restClientBuilder -> {
+            SimpleClientHttpRequestFactory factory = new SimpleClientHttpRequestFactory();
+            factory.setConnectTimeout((int) Duration.ofMinutes(10).toMillis());
+            factory.setReadTimeout((int) Duration.ofMinutes(10).toMillis());
+            restClientBuilder.defaultHeader(HttpHeaders.ACCEPT_ENCODING, "gzip, deflate");
+            restClientBuilder.requestFactory(factory);
+        };
+    }
+}

src/main/java/me/pacphi/config/Http.java

@@ -2,11 +2,11 @@
 
 import io.micrometer.observation.ObservationRegistry;
 import org.apache.commons.lang3.StringUtils;
-import org.springframework.ai.autoconfigure.openai.OpenAiChatProperties;
-import org.springframework.ai.autoconfigure.openai.OpenAiConnectionProperties;
 import org.springframework.ai.chat.client.ChatClient;
 import org.springframework.ai.chat.client.advisor.SimpleLoggerAdvisor;
 import org.springframework.ai.model.SimpleApiKey;
+import org.springframework.ai.model.openai.autoconfigure.OpenAiChatProperties;
+import org.springframework.ai.model.openai.autoconfigure.OpenAiConnectionProperties;
 import org.springframework.ai.model.tool.DefaultToolCallingManager;
 import org.springframework.ai.openai.OpenAiChatModel;
 import org.springframework.ai.openai.OpenAiChatOptions;
@@ -59,23 +59,23 @@ public Map<String, ChatClient> chatClients(
 
         return multiChatProperties.getOptions().getModels().stream().collect(
                 Collectors.toMap(
-                    model -> model,
-                    model -> {
-                        OpenAiChatOptions chatOptions = OpenAiChatOptions.fromOptions(chatProperties.getOptions());
-                        chatOptions.setModel(model);
-                        OpenAiChatModel openAiChatModel = new OpenAiChatModel(
-                                openAiApi,
-                                chatProperties.getOptions(),
-                                DefaultToolCallingManager.builder().observationRegistry(observationRegistry).build(),
-                                retryTemplate,
-                                observationRegistry
-                        );
-                        // Create ChatClient with similar configuration to original service
-                        return ChatClient.builder(openAiChatModel)
-                                .defaultAdvisors(
-                                        new SimpleLoggerAdvisor())
-                                .build();
-                    }
+                        model -> model,
+                        model -> {
+                            OpenAiChatOptions chatOptions = OpenAiChatOptions.fromOptions(chatProperties.getOptions());
+                            chatOptions.setModel(model);
+                            OpenAiChatModel openAiChatModel = new OpenAiChatModel(
+                                    openAiApi,
+                                    chatOptions,
+                                    DefaultToolCallingManager.builder().observationRegistry(observationRegistry).build(),
+                                    retryTemplate,
+                                    observationRegistry
+                            );
+                            // Create ChatClient with similar configuration to original service
+                            return ChatClient.builder(openAiChatModel)
+                                    .defaultAdvisors(
+                                            new SimpleLoggerAdvisor())
+                                    .build();
+                        }
                 )
         );
     }

src/main/java/me/pacphi/config/MultiChat.java

@@ -10,6 +10,7 @@ public record MultiChatResponse(
         boolean success,
         String errorMessage,
         Integer promptTokens,
+        Integer completionTokens,
         Integer totalTokens,
         String responseTime
 ) {
@@ -32,10 +33,10 @@ private static String getFormattedResponseTime(long responseTimeMillis) {
     }
 
     public static MultiChatResponse success(String model, String response, Usage usage, long responseTime) {
-        return new MultiChatResponse(model, response, true, null, usage.getPromptTokens(), usage.getTotalTokens(), getFormattedResponseTime(responseTime));
+        return new MultiChatResponse(model, response, true, null, usage.getPromptTokens(), usage.getCompletionTokens(), usage.getTotalTokens(), getFormattedResponseTime(responseTime));
     }
 
     public static MultiChatResponse failure(String model, String error) {
-        return new MultiChatResponse(model, null, false, error, null, null, null);
+        return new MultiChatResponse(model, null, false, error, null, null, null, null);
     }
 }