QianFan Chat

Spring AI supports the various AI language models from QianFan. You can interact with QianFan language models and create a multilingual conversational assistant based on QianFan models.

Prerequisites

You will need to create an API with QianFan to access QianFan language models.

Create an account at QianFan registration page and generate the token on the API Keys page. The Spring AI project defines a configuration property named spring.ai.qianfan.api-key and spring.ai.qianfan.secret-key. you should set to the value of the API Key and Secret Key obtained from API Keys page. Exporting an environment variable is one way to set that configuration property:

export SPRING_AI_QIANFAN_API_KEY=<INSERT KEY HERE>

Add Repositories and BOM

Spring AI artifacts are published in Spring Milestone and Snapshot repositories. Refer to the Repositories section to add these repositories to your build system.

To help with dependency management, Spring AI provides a BOM (bill of materials) to ensure that a consistent version of Spring AI is used throughout the entire project. Refer to the Dependency Management section to add the Spring AI BOM to your build system.

Auto-configuration

Spring AI provides Spring Boot auto-configuration for the QianFan Chat Client. To enable it add the following dependency to your project’s Maven pom.xml file:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-qianfan-spring-boot-starter</artifactId>
</dependency>

or to your Gradle build.gradle build file.

dependencies {
    implementation 'org.springframework.ai:spring-ai-qianfan-spring-boot-starter'
}
Refer to the Dependency Management section to add the Spring AI BOM to your build file.

Chat Properties

Retry Properties

The prefix spring.ai.retry is used as the property prefix that lets you configure the retry mechanism for the QianFan Chat client.

Property Description Default

spring.ai.retry.max-attempts

Maximum number of retry attempts.

10

spring.ai.retry.backoff.initial-interval

Initial sleep duration for the exponential backoff policy.

2 sec.

spring.ai.retry.backoff.multiplier

Backoff interval multiplier.

5

spring.ai.retry.backoff.max-interval

Maximum backoff duration.

3 min.

spring.ai.retry.on-client-errors

If false, throw a NonTransientAiException, and do not attempt retry for 4xx client error codes

false

spring.ai.retry.exclude-on-http-codes

List of HTTP status codes that should not trigger a retry (e.g. to throw NonTransientAiException).

empty

spring.ai.retry.on-http-codes

List of HTTP status codes that should trigger a retry (e.g. to throw TransientAiException).

empty

Connection Properties

The prefix spring.ai.qianfan is used as the property prefix that lets you connect to QianFan.

Property Description Default

spring.ai.qianfan.base-url

The URL to connect to

api.qianfan.chat

spring.ai.qianfan.api-key

The API Key

-

spring.ai.qianfan.secret-key

The Secret Key

-

Configuration Properties

The prefix spring.ai.qianfan.chat is the property prefix that lets you configure the chat client implementation for QianFan.

Property Description Default

spring.ai.qianfan.chat.enabled

Enable QianFan chat client.

true

spring.ai.qianfan.chat.base-url

Optional overrides the spring.ai.qianfan.base-url to provide chat specific url

api.qianfan.chat

spring.ai.qianfan.chat.api-key

Optional overrides the spring.ai.qianfan.api-key to provide chat specific api-key

-

spring.ai.qianfan.chat.secret-key

Optional overrides the spring.ai.qianfan.secret-key to provide chat specific secret-key

-

spring.ai.qianfan.chat.options.model

This is the QianFan Chat model to use

abab5.5-chat (the abab5.5s-chat, abab5.5-chat, and abab6-chat point to the latest model versions)

spring.ai.qianfan.chat.options.maxTokens

The maximum number of tokens to generate in the chat completion. The total length of input tokens and generated tokens is limited by the model’s context length.

-

spring.ai.qianfan.chat.options.temperature

The sampling temperature to use that controls the apparent creativity of generated completions. Higher values will make output more random while lower values will make results more focused and deterministic. It is not recommended to modify temperature and top_p for the same completions request as the interaction of these two settings is difficult to predict.

0.7

spring.ai.qianfan.chat.options.topP

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or temperature but not both.

1.0

spring.ai.qianfan.chat.options.presencePenalty

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model’s likelihood to talk about new topics.

0.0f

spring.ai.qianfan.chat.options.frequencyPenalty

Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model’s likelihood to repeat the same line verbatim.

0.0f

spring.ai.qianfan.chat.options.stop

The model will stop generating characters specified by stop, and currently only supports a single stop word in the format of ["stop_word1"]

-
You can override the common spring.ai.qianfan.base-url, spring.ai.qianfan.chat.api-key and spring.ai.qianfan.chat.secret-key for the ChatClient implementations. The spring.ai.qianfan.chat.base-url, spring.ai.qianfan.chat.api-key and spring.ai.qianfan.chat.secret-key properties if set take precedence over the common properties. This is useful if you want to use different QianFan accounts for different models and different model endpoints.
All properties prefixed with spring.ai.qianfan.chat.options can be overridden at runtime by adding a request specific Runtime Options to the Prompt call.

Runtime Options

The QianFanChatOptions.java provides model configurations, such as the model to use, the temperature, the frequency penalty, etc.

On start-up, the default options can be configured with the QianFanChatModel(api, options) constructor or the spring.ai.qianfan.chat.options.* properties.

At run-time you can override the default options by adding new, request specific, options to the Prompt call. For example to override the default model and temperature for a specific request:

ChatResponse response = chatClient.call(
    new Prompt(
        "Generate the names of 5 famous pirates.",
        QianFanChatOptions.builder()
            .withModel(QianFanApi.ChatModel.ERNIE_Speed_8K.getValue())
            .withTemperature(0.5f)
        .build()
    ));
In addition to the model specific QianFanChatOptions you can use a portable ChatOptions instance, created with the ChatOptionsBuilder#builder().

Sample Controller

Create a new Spring Boot project and add the spring-ai-qianfan-spring-boot-starter to your pom (or gradle) dependencies.

Add a application.properties file, under the src/main/resources directory, to enable and configure the QianFan Chat client:

spring.ai.qianfan.api-key=YOUR_API_KEY
spring.ai.qianfan.secret-key=YOUR_SECRET_KEY
spring.ai.qianfan.chat.options.model=ernie_speed
spring.ai.qianfan.chat.options.temperature=0.7
replace the api-key and secret-key with your QianFan credentials.

This will create a QianFanChatModel implementation that you can inject into your class. Here is an example of a simple @Controller class that uses the chat client for text generations.

@RestController
public class ChatController {

    private final QianFanChatModel chatClient;

    @Autowired
    public ChatController(QianFanChatModel chatClient) {
        this.chatClient = chatClient;
    }

    @GetMapping("/ai/generate")
    public Map generate(@RequestParam(value = "message", defaultValue = "Tell me a joke") String message) {
        return Map.of("generation", chatClient.call(message));
    }

    @GetMapping("/ai/generateStream")
	public Flux<ChatResponse> generateStream(@RequestParam(value = "message", defaultValue = "Tell me a joke") String message) {
        var prompt = new Prompt(new UserMessage(message));
        return chatClient.stream(prompt);
    }
}

Manual Configuration

The QianFanChatModel implements the ChatClient and StreamingChatClient and uses the Low-level QianFanApi Client to connect to the QianFan service.

Add the spring-ai-qianfan dependency to your project’s Maven pom.xml file:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-qianfan</artifactId>
</dependency>

or to your Gradle build.gradle build file.

dependencies {
    implementation 'org.springframework.ai:spring-ai-qianfan'
}
Refer to the Dependency Management section to add the Spring AI BOM to your build file.

Next, create a QianFanChatModel and use it for text generations:

var qianFanApi = new QianFanApi(System.getenv("QIANFAN_API_KEY"), System.getenv("QIANFAN_SECRET_KEY"));

var chatClient = new QianFanChatModel(qianFanApi, QianFanChatOptions.builder()
                .withModel(QianFanApi.ChatModel.ERNIE_Speed_8K.getValue())
                .withTemperature(0.4f)
                .withMaxTokens(200)
                .build());

ChatResponse response = chatClient.call(
    new Prompt("Generate the names of 5 famous pirates."));

// Or with streaming responses
Flux<ChatResponse> streamResponse = chatClient.stream(
    new Prompt("Generate the names of 5 famous pirates."));

The QianFanChatOptions provides the configuration information for the chat requests. The QianFanChatOptions.Builder is fluent options builder.

Low-level QianFanApi Client

The QianFanApi provides is lightweight Java client for QianFan API.

Here is a simple snippet how to use the api programmatically:

String systemMessage = "Your name is QianWen";

QianFanApi qianFanApi =
    new QianFanApi(System.getenv("QIANFAN_API_KEY"), System.getenv("QIANFAN_SECRET_KEY"));

ChatCompletionMessage chatCompletionMessage =
    new ChatCompletionMessage("Hello world", Role.USER);

// Sync request
ResponseEntity<ChatCompletion> response = qianFanApi.chatCompletionEntity(
    new ChatCompletionRequest(List.of(chatCompletionMessage), systemMessage, QianFanApi.ChatModel.ERNIE_Speed_8K.getValue(), 0.7f, false));

// Streaming request
Flux<ChatCompletionChunk> streamResponse = qianFanApi.chatCompletionStream(
        new ChatCompletionRequest(List.of(chatCompletionMessage), systemMessage, QianFanApi.ChatModel.ERNIE_Speed_8K.getValue(), 0.7f, true));

Follow the QianFanApi.java's JavaDoc for further information.

QianFanApi Samples

  • The QianFanApiIT.java test provides some general examples how to use the lightweight library.