Integrate DeepSeek with Spring AI: Step‑by‑Step Guide for Java Developers

Spring AI is a project in the Spring ecosystem that integrates AI into Spring applications, supporting major providers like Anthropic, OpenAI, Microsoft, Amazon, Google, and Ollama.

Spring AI connects your data and APIs with AI models.

Chat

Embedding

Text‑to‑Image

Audio‑to‑Text

Text‑to‑Audio

DeepSeek, developed by the Chinese team DeepSeek, is a multimodal large model with strong reasoning and code generation abilities, offering low cost and high performance.

Low cost (no special hardware, open‑source, simple usage).

High performance (strong inference, accurate answers).

Integration Steps

1. Environment Preparation

JDK 17 or higher

Maven or Gradle

DeepSeek API Key (obtain from https://platform.deepseek.com/usage)

2. Create Spring Boot Project

Use Spring Initializr to create a Spring Boot 3.2.x project.

3. Add Dependencies

<dependencies>
  <dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-web</artifactId>
  </dependency>
  <dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
  </dependency>
</dependencies>

4. Configuration

# Required
spring.ai.openai.api-key=your-api-key
spring.ai.openai.base-url=https://api.deepseek.com
# Model selection (example uses chat model)
spring.ai.openai.chat.options.model=deepseek-chat

5. Write Code

Create a controller that uses DeepSeekClient for chat and streaming responses.

import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.web.bind.annotation.*;

@RestController
@RequestMapping("/api/chat")
public class ChatController {

    @Autowired
    private DeepSeekClient deepSeekClient;

    @PostMapping
    public String chat(@RequestBody String message) {
        return deepSeekClient.chatCompletion(message).getOutput().getContent();
    }

    @GetMapping(value = "/stream", produces = "text/event-stream")
    public Flux<String> chatStream(@RequestParam String message) {
        return deepSeekClient.chatFluxCompletion(message)
                .map(response -> response.getOutput().getContent());
    }
}

6. Streaming Output Considerations

Streaming reduces perceived latency by sending partial results; the backend must implement server‑sent events and the frontend must handle them.