Using Spring AI to Access Deepseek Through Alibaba Cloud Model Studio

One day when I check the maven BOM of Spring-AI, I found a Springboot starter for Spring AI to access Dashscope, which was the former name of Alibaba Cloud Model Studio.

But the doc of Spring AI didn't announce that, so, let's learn to use it.

Spring AI need your jdk version >= 17

1. Apply a Api-key on Alibaba Cloud Model Studio

Alibaba Cloud Model Studio - Alibaba Cloud

A one-stop generative AI platform to build intelligent applications that understand your business, based on Qwen and other popular models

alibabacloud.com

2. Import Spring-AI

This is a new Springboot maven project

• Add Spring Milestone and Snapshot Repositories

You will need to add Spring milestone and snapshot repositories to your build system. Because Spring didn't publish it to maven center repository.

<repositories>
    <repository>
        <id>spring-milestones</id>
        <name>Spring Milestones</name>
        <url>https://repo.spring.io/milestone</url>
        <snapshots>
            <enabled>false</enabled>
        </snapshots>
    </repository>
    <repository>
        <id>spring-snapshots</id>
        <name>Spring Snapshots</name>
        <url>https://repo.spring.io/snapshot</url>
        <releases>
            <enabled>false</enabled>
        </releases>
    </repository>
</repositories>

• Import Spring AI BOM to Manage the Version of Spring AI

<dependencyManagement>
    <dependencies>
        <dependency>
            <groupId>group.springframework.ai</groupId>
            <artifactId>spring-ai-bom</artifactId>
            <version>${spring-ai.version}</version>
            <type>pom</type>
            <scope>import</scope>
        </dependency>
        <dependency>
            <groupId>org.springframework.boot</groupId>
            <artifactId>spring-boot-dependencies</artifactId>
            <version>3.3.4</version>
            <type>pom</type>
            <scope>import</scope>
        </dependency>
    </dependencies>
</dependencyManagement>

• Add spring-ai-dashscope-spring-boot-starter and Some Assistant Package

<dependencies>
    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-web</artifactId>
    </dependency>
    <dependency>
        <groupId>org.projectlombok</groupId>
        <artifactId>lombok</artifactId>
        <optional>true</optional>
    </dependency>
    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-test</artifactId>
        <scope>test</scope>
    </dependency>
    <dependency>
        <groupId>group.springframework.ai</groupId>
        <artifactId>spring-ai-dashscope-spring-boot-starter</artifactId>
    </dependency>
</dependencies>

3. Write application.yaml

spring:
  ai:
    dashscope:
      # DashScope API Key
      api-key: # TODO
      chat:
        options:
          # choose the model you want to use
          model: deepseek-v3
          # temperature is the probability of the model to generate a new word, 0.7 is the default value
          temperature: 0.7

logging:
  level:
    root: info
    # if you want to check the log of Spring AI, you need enable this config
#    org.springframework: debug
server:
  port: 9090

The value of temperature you can fellow the advice of deepseek

4. Register ChatClient as a Spring Bean

@Configuration
public class SpringAiConfig {

    @Autowired
    private ChatModel chatModel;

    @Bean
    public ChatClient chatClient() {
        return ChatClient.builder(chatModel)
                // Set default system prompt
                .defaultSystem("""
                        Imitate Max Verstappen's tone
                        date of today is {current_date}
                        """)
                .build();
    }
}

Then you can use ChatClient to access AI model

5. Test ChatClient

I create a controller to test ChatClient, All test will based on web request to access.

@RestController
@RequiredArgsConstructor
@RequestMapping("/ai")
@Slf4j
public class AiController {

    private final ChatClient chatClient;

}

Let's start Spring AI with a normal chat

@GetMapping("/chat")
public String chat(@RequestParam(value = "msg") String msg) {
    log.info("msg: {}", msg);
    return chatClient.prompt()
            .user(msg)
            // config default system's current date
            .system(promptSystemSpec -> promptSystemSpec.param("current_date", LocalDateTime.now()))
            .call()
            .content();
}

Start the project and test it using postman

How amazing it is!!!🎉🎉🎉

---

If you want to learn the advance of Spring AI, follow me and I will publish it later.