快捷導(dǎo)航

Springboot整合Ollama實(shí)現(xiàn)調(diào)用本地多模型

更新時(shí)間：2026年07月12日 13:59:36 作者：Listen·Rain

想知道如何用ollama快速安裝和調(diào)用大模型嗎,本文帶你從ollama官網(wǎng)搜索模型、使用powershell命令一鍵安裝,到springboot項(xiàng)目中引入依賴、配置參數(shù)、編寫服務(wù),最后通過controller測(cè)試接口,輕松實(shí)現(xiàn)ollama大模型的本地部署和api調(diào)用,需要的朋友可以參考下

1.安裝相應(yīng)的大模型

可以在ollama官網(wǎng)搜索相應(yīng)的大模型，直接copy相應(yīng)命令，在powershell窗口執(zhí)行ollama命令

查看命令：

ollama list

2.pom.xml

引入Ollama依賴

<dependency>
			<groupId>org.springframework.ai</groupId>
			<artifactId>spring-ai-starter-model-ollama</artifactId>
			<version>1.0.0</version>
		</dependency>

3.application.properties

server.port: 8088
server.servlet.encoding.charset=UTF-8
server.servlet.encoding.enabled=true
server.servlet.encoding.force=true
spring.ai.ollama.base-url=http://localhost:11434

4.OllamaConfig

@Configuration
public class OllamaConfig {

	private static final String QWEN = "qwen3:0.6b";
	private static final String DEEPSEEK = "deepseek-r1:1.5b";

	@Bean(name = "qwenChatClient")
	public ChatClient qwenChatClient(ChatClient.Builder builder) {
		OllamaOptions ollamaOptions = OllamaOptions.builder().build();
		ollamaOptions.setModel(QWEN);
		return builder.defaultOptions(ollamaOptions).build();
	}

	@Bean(name = "deepseekChatClient")
	public ChatClient deepseekChatClient(ChatClient.Builder builder) {
		OllamaOptions ollamaOptions = OllamaOptions.builder().build();
		ollamaOptions.setModel(DEEPSEEK);
		return builder.defaultOptions(ollamaOptions).build();
	}
}

5.OllamaService

package com.example.demoredisstack.service;

import org.springframework.ai.chat.client.ChatClient;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;
import reactor.core.publisher.Flux;

@Service
public class OllamaService {

	@Autowired
	private ChatClient qwenChatClient;

	@Autowired
	private ChatClient deepseekChatClient;


	public String qwenChat(String message) {
		return qwenChatClient
				.prompt()       // 開始構(gòu)建提示詞
				.user(message)  // 設(shè)置用戶消息
				.call()         // 發(fā)起調(diào)用
				.content();     // 直接獲取字符串內(nèi)容
	}

	public Flux<String> qwenChatStream(String message) {
		return qwenChatClient
				.prompt()
				.user(message)
				.stream()
				.content();
	}

	public String deepseekChat(String message) {
		return deepseekChatClient
				.prompt()       // 開始構(gòu)建提示詞
				.user(message)  // 設(shè)置用戶消息
				.call()         // 發(fā)起調(diào)用
				.content();     // 直接獲取字符串內(nèi)容
	}

	public Flux<String> deepseekChatStream(String message) {
		return deepseekChatClient
				.prompt()
				.user(message)
				.stream()
				.content();
	}
}

6. TestOllamaController

package com.example.demoredisstack.controller;

import com.example.demoredisstack.service.OllamaService;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
import reactor.core.publisher.Flux;

@RestController
@RequestMapping("/test/ollama")
public class TestOllamaController {

	@Autowired
	private OllamaService ollamaService;

	@GetMapping("/qwen")
	public String qwenChat(@RequestParam(value = "message", defaultValue = "你好") String message) {
		return ollamaService.qwenChat(message);
	}

	@GetMapping("/qwenStream")
	public Flux<String> qwenChatSteam(@RequestParam(value = "message", defaultValue = "你好") String message) {
		return ollamaService.qwenChatStream(message);
	}

	@GetMapping("/deepseek")
	public String deepseekChat(@RequestParam(value = "message", defaultValue = "你好") String message) {
		return ollamaService.deepseekChat(message);
	}

	@GetMapping("/deepseekStream")
	public Flux<String> deepseekChatStream(@RequestParam(value = "message", defaultValue = "你好") String message) {
		return ollamaService.deepseekChatStream(message);
	}
}