The Spring AI Bean Output Converter Using Low Level Chat Mode is provides a fluent API which is sufficient for simple Java Beans responses, but if we require more complex target class structure output from LLM AI Model, such as a List or Map of the Target Class. The Spring AI Generic Bean Types Using Low Level Chat Model uses the ParameterizedTypeReference and allows to specify more complex target class structures.
package com.example.springai.controller;
import com.example.springai.entity.Country;
import org.springframework.ai.chat.model.ChatModel;
import org.springframework.ai.chat.prompt.PromptTemplate;
import org.springframework.ai.converter.BeanOutputConverter;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.core.ParameterizedTypeReference;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
import java.util.Map;
@RestController
public class SpringAiController {
@Autowired
private ChatModel chatModel;
@GetMapping("/beanOutputParser")
Country beanOutputParser(@RequestParam(value = "letter", defaultValue = "a") String letter) {
var template = """
Give me a country starts with {letter} and its capital.
{format}
""";
var outputConverter = new BeanOutputConverter<Country>(new ParameterizedTypeReference<>() {
});
var prompt = new PromptTemplate(template, Map.of("letter", letter, "format", outputConverter.getFormat())).create();
var generation = chatModel.call(prompt).getResult();
return outputConverter.convert(generation.getOutput().getContent());
}
}
package com.example.springai.entity;
public record Country(String country, String capital) {
}
package com.example.springai;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
@SpringBootApplication
public class SpringAiApplication {
public static void main(String[] args) {
SpringApplication.run(SpringAiApplication.class, args);
}
}
spring.application.name=SpringAi
spring.docker.compose.lifecycle-management=start-only
spring.threads.virtual.enabled=true
# The default Ollama Model in Spring Ai is mistral, but it can be changed by setting the below property. make sure to download the same model in entrypoint.sh file
#spring.ai.ollama.chat.options.model=llama3.1
# If running the Ollama Docker Instance separately, then set this property
spring.docker.compose.enabled=false
services:
ollama-model:
image: ollama/ollama:latest
container_name: ollama_container
ports:
- 11434:11434/tcp
healthcheck:
test: ollama --version || exit 1
command: serve
volumes:
- ./ollama/ollama:/root/.ollama
- ./entrypoint.sh:/entrypoint.sh
pull_policy: missing
tty: true
restart: no
entrypoint: [ "/usr/bin/bash", "/entrypoint.sh" ]
open-webui:
image: ghcr.io/open-webui/open-webui:main
container_name: open_webui_container
environment:
WEBUI_AUTH: false
ports:
- "8081:8080"
extra_hosts:
- "host.docker.internal:host-gateway"
volumes:
- open-webui:/app/backend/data
restart: no
volumes:
open-webui:
#!/bin/bash
# Start Ollama in the background.
/bin/ollama serve &
# Record Process ID.
pid=$!
# Pause for Ollama to start.
sleep 5
# The default Ollama Model in Spring Ai is mistral, but it can be changed in the applications property file. Make sure to download the same Model here
echo "🔴 Retrieve LLAMA3 model..."
ollama pull mistral
echo "🟢 Done!"
# Wait for the Ollama process to finish.
wait $pid
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://maven.apache.org/POM/4.0.0"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-parent</artifactId>
<version>3.3.2</version>
<relativePath/>
</parent>
<groupId>com.example.springai</groupId>
<artifactId>generic_bean_types_using_low_level_chatmodel_api</artifactId>
<version>0.0.1-SNAPSHOT</version>
<name>Generic Bean Types using Low Level ChatModel API</name>
<description>Demo project for Spring Boot</description>
<properties>
<java.version>21</java.version>
<spring-ai.version>1.0.0-SNAPSHOT</spring-ai.version>
</properties>
<dependencies>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.ai</groupId>
<artifactId>spring-ai-ollama-spring-boot-starter</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-docker-compose</artifactId>
<scope>runtime</scope>
<optional>true</optional>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-test</artifactId>
<scope>test</scope>
</dependency>
</dependencies>
<dependencyManagement>
<dependencies>
<dependency>
<groupId>org.springframework.ai</groupId>
<artifactId>spring-ai-bom</artifactId>
<version>${spring-ai.version}</version>
<type>pom</type>
<scope>import</scope>
</dependency>
</dependencies>
</dependencyManagement>
<build>
<plugins>
<plugin>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-maven-plugin</artifactId>
<configuration>
<mainClass>com.example.springai.SpringAiApplication</mainClass>
<excludes>
<exclude>
<groupId>org.projectlombok</groupId>
<artifactId>lombok</artifactId>
</exclude>
</excludes>
</configuration>
</plugin>
</plugins>
</build>
<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>
</project>
run the curl to see the Spring AI Generic Bean Types Using Low Level Chat Model
curl --location 'localhost:8080/beanOutputParser'
