MiniMax Chat
Spring AI 支持 MiniMax 的各种 AI 语言模型。您可以与 MiniMax 语言模型交互，并基于 MiniMax 模型创建多语言对话助手。
Spring AI supports the various AI language models from MiniMax. You can interact with MiniMax language models and create a multilingual conversational assistant based on MiniMax models.
Prerequisites
您需要使用 MiniMax 创建一个 API 以访问 MiniMax 语言模型。
You will need to create an API with MiniMax to access MiniMax language models.
在 MiniMax registration page 创建一个帐户，并在 API Keys page 生成令牌。
Create an account at MiniMax registration page and generate the token on the API Keys page.
Spring AI 项目定义了一个名为 spring.ai.minimax.api-key 的配置属性，您应该将其设置为从 API 密钥页面获取的 API Key 的值。
The Spring AI project defines a configuration property named spring.ai.minimax.api-key that you should set to the value of the API Key obtained from the API Keys page.
你可以在` application.properties `文件中设置此配置属性：
You can set this configuration property in your application.properties file:
spring.ai.minimax.api-key=<your-minimax-api-key>为了在处理API密钥等敏感信息时增强安全性，你可以使用Spring表达式语言（SpEL）引用环境变量：
For enhanced security when handling sensitive information like API keys, you can use Spring Expression Language (SpEL) to reference an environment variable:
# In application.yml
spring:
ai:
minimax:
api-key: ${MINIMAX_API_KEY}# In your environment or .env file
export MINIMAX_API_KEY=<your-minimax-api-key>你也可以在应用程序代码中以编程方式设置此配置：
You can also set this configuration programmatically in your application code:
// Retrieve API key from a secure source or environment variable
String apiKey = System.getenv("MINIMAX_API_KEY");Add Repositories and BOM
Spring AI 工件发布在 Maven Central 和 Spring Snapshot 存储库中。请参阅“添加 Spring AI 仓库”部分，将这些仓库添加到您的构建系统。
Spring AI artifacts are published in Maven Central and Spring Snapshot repositories. Refer to the Artifact Repositories section to add these repositories to your build system.
为了帮助进行依赖项管理，Spring AI 提供了一个 BOM（物料清单）以确保在整个项目中使用一致版本的 Spring AI。有关将 Spring AI BOM 添加到你的构建系统的说明，请参阅 Dependency Management 部分。
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 自动配置、启动器模块的工件名称发生了重大变化。请参阅 upgrade notes 以获取更多信息。 There has been a significant change in the Spring AI auto-configuration, starter modules' artifact names. Please refer to the upgrade notes for more information. |
Spring AI 为 MiniMax 聊天客户端提供了 Spring Boot 自动配置。要启用它，请将以下依赖项添加到您项目的 Maven pom.xml 文件中：
Spring AI provides Spring Boot auto-configuration for the MiniMax 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-starter-model-minimax</artifactId>
</dependency>或添加到 Gradle build.gradle 构建文件中。
or to your Gradle build.gradle build file.
dependencies {
implementation 'org.springframework.ai:spring-ai-starter-model-minimax'
}
|
|
Refer to the Dependency Management section to add the Spring AI BOM to your build file. |
Chat Properties
Retry Properties
前缀 spring.ai.retry 用作属性前缀，允许您配置 MiniMax 聊天模型的重试机制。
The prefix spring.ai.retry is used as the property prefix that lets you configure the retry mechanism for the MiniMax chat model.
| 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 |
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
前缀 spring.ai.minimax 用作属性前缀，允许您连接到 MiniMax。
The prefix spring.ai.minimax is used as the property prefix that lets you connect to MiniMax.
| Property | Description | Default |
|---|---|---|
spring.ai.minimax.base-url |
The URL to connect to |
[role="bare"]https://api.minimax.chat |
spring.ai.minimax.api-key |
The API Key |
- |
Configuration Properties
|
聊天自动配置的启用和禁用现在通过前缀为 Enabling and disabling of the chat auto-configurations are now configured via top level properties with the prefix 要启用，spring.ai.model.chat=minimax（默认启用） To enable, spring.ai.model.chat=minimax (It is enabled by default) 要禁用，spring.ai.model.chat=none（或任何不匹配 minimax 的值） To disable, spring.ai.model.chat=none (or any value which doesn’t match minimax) 此更改旨在允许配置多个模型。 This change is done to allow configuration of multiple models. |
前缀 spring.ai.minimax.chat 是属性前缀，允许您配置 MiniMax 的聊天模型实现。
The prefix spring.ai.minimax.chat is the property prefix that lets you configure the chat model implementation for MiniMax.
| Property | Description | Default |
|---|---|---|
spring.ai.minimax.chat.enabled (Removed and no longer valid) |
Enable MiniMax chat model. |
true |
spring.ai.model.chat |
Enable MiniMax chat model. |
minimax |
spring.ai.minimax.chat.base-url |
Optional overrides the spring.ai.minimax.base-url to provide chat specific url |
[role="bare"]https://api.minimax.chat |
spring.ai.minimax.chat.api-key |
Optional overrides the spring.ai.minimax.api-key to provide chat specific api-key |
- |
spring.ai.minimax.chat.options.model |
This is the MiniMax Chat model to use |
|
spring.ai.minimax.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.minimax.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.minimax.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.minimax.chat.options.n |
How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Default value is 1 and cannot be greater than 5. Specifically, when the temperature is very small and close to 0, we can only return 1 result. If n is already set and>1 at this time, service will return an illegal input parameter (invalid_request_error) |
1 |
spring.ai.minimax.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.minimax.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.minimax.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"] |
- |
|
您可以为 |
|
You can override the common |
|
所有前缀为 |
|
All properties prefixed with |
Runtime Options
MiniMaxChatOptions.java 提供模型配置，例如要使用的模型、温度、频率惩罚等。
The MiniMaxChatOptions.java provides model configurations, such as the model to use, the temperature, the frequency penalty, etc.
启动时，可以使用 MiniMaxChatModel(api, options) 构造函数或 spring.ai.minimax.chat.options.* 属性配置默认选项。
On start-up, the default options can be configured with the MiniMaxChatModel(api, options) constructor or the spring.ai.minimax.chat.options.* properties.
在运行时，可以通过向 Prompt 调用中添加新的请求特定选项来覆盖默认选项。例如，覆盖特定请求的默认模型和温度：
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 = chatModel.call(
new Prompt(
"Generate the names of 5 famous pirates.",
MiniMaxChatOptions.builder()
.model(MiniMaxApi.ChatModel.ABAB_6_5_S_Chat.getValue())
.temperature(0.5)
.build()
));|
除了模型特定的 MiniMaxChatOptions ，您还可以使用通过 ChatOptionsBuilder#builder() 创建的便携式 ChatOptions 实例。 |
|
In addition to the model specific MiniMaxChatOptions you can use a portable ChatOptions instance, created with the ChatOptionsBuilder#builder(). |
Sample Controller
Create 一个新的 Spring Boot 项目，并将 spring-ai-starter-model-minimax 添加到您的 pom（或 gradle）依赖项中。
Create a new Spring Boot project and add the spring-ai-starter-model-minimax to your pom (or gradle) dependencies.
在 src/main/resources 目录下添加一个 application.properties 文件，以启用和配置 MiniMax 聊天模型：
Add a application.properties file, under the src/main/resources directory, to enable and configure the MiniMax chat model:
spring.ai.minimax.api-key=YOUR_API_KEY
spring.ai.minimax.chat.options.model=abab6.5g-chat
spring.ai.minimax.chat.options.temperature=0.7|
将 |
|
replace the |
这将创建一个 MiniMaxChatModel 实现，您可以将其注入到您的类中。以下是一个简单的 @Controller 类的示例，该类使用聊天模型进行文本生成。
This will create a MiniMaxChatModel implementation that you can inject into your class.
Here is an example of a simple @Controller class that uses the chat model for text generations.
@RestController
public class ChatController {
private final MiniMaxChatModel chatModel;
@Autowired
public ChatController(MiniMaxChatModel chatModel) {
this.chatModel = chatModel;
}
@GetMapping("/ai/generate")
public Map generate(@RequestParam(value = "message", defaultValue = "Tell me a joke") String message) {
return Map.of("generation", this.chatModel.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 this.chatModel.stream(prompt);
}
}Manual Configuration
The MiniMaxChatModel 实现了 ChatModel 和 StreamingChatModel ，并使用 Low-level MiniMaxApi Client 连接到 MiniMax 服务。
The MiniMaxChatModel implements the ChatModel and StreamingChatModel and uses the Low-level MiniMaxApi Client to connect to the MiniMax service.
将 spring-ai-minimax 依赖项添加到您项目的 Maven pom.xml 文件中：
Add the spring-ai-minimax dependency to your project’s Maven pom.xml file:
<dependency>
<groupId>org.springframework.ai</groupId>
<artifactId>spring-ai-minimax</artifactId>
</dependency>或添加到 Gradle build.gradle 构建文件中。
or to your Gradle build.gradle build file.
dependencies {
implementation 'org.springframework.ai:spring-ai-minimax'
}
|
|
Refer to the Dependency Management section to add the Spring AI BOM to your build file. |
接下来，创建一个 MiniMaxChatModel 并将其用于文本生成：
Next, create a MiniMaxChatModel and use it for text generations:
var miniMaxApi = new MiniMaxApi(System.getenv("MINIMAX_API_KEY"));
var chatModel = new MiniMaxChatModel(this.miniMaxApi, MiniMaxChatOptions.builder()
.model(MiniMaxApi.ChatModel.ABAB_6_5_S_Chat.getValue())
.temperature(0.4)
.maxTokens(200)
.build());
ChatResponse response = this.chatModel.call(
new Prompt("Generate the names of 5 famous pirates."));
// Or with streaming responses
Flux<ChatResponse> streamResponse = this.chatModel.stream(
new Prompt("Generate the names of 5 famous pirates."));MiniMaxChatOptions 提供聊天请求的配置信息。 MiniMaxChatOptions.Builder 是一个流式选项构建器。
The MiniMaxChatOptions provides the configuration information for the chat requests.
The MiniMaxChatOptions.Builder is fluent options builder.
Low-level MiniMaxApi Client
MiniMaxApi 为 MiniMax API 提供了轻量级 Java 客户端。
The MiniMaxApi provides is lightweight Java client for MiniMax API.
下面是一个简单的片段，说明如何以编程方式使用 API：
Here is a simple snippet how to use the api programmatically:
MiniMaxApi miniMaxApi =
new MiniMaxApi(System.getenv("MINIMAX_API_KEY"));
ChatCompletionMessage chatCompletionMessage =
new ChatCompletionMessage("Hello world", Role.USER);
// Sync request
ResponseEntity<ChatCompletion> response = this.miniMaxApi.chatCompletionEntity(
new ChatCompletionRequest(List.of(this.chatCompletionMessage), MiniMaxApi.ChatModel.ABAB_6_5_S_Chat.getValue(), 0.7f, false));
// Streaming request
Flux<ChatCompletionChunk> streamResponse = this.miniMaxApi.chatCompletionStream(
new ChatCompletionRequest(List.of(this.chatCompletionMessage), MiniMaxApi.ChatModel.ABAB_6_5_S_Chat.getValue(), 0.7f, true));有关更多信息，请遵循 MiniMaxApi.java 的 JavaDoc。
Follow the MiniMaxApi.java's JavaDoc for further information.
WebSearch chat
MiniMax 模型支持网页搜索功能。网页搜索功能允许您搜索网页以获取信息，并将结果返回到聊天响应中。
The MiniMax model supported the web search feature. The web search feature allows you to search the web for information and return the results in the chat response.
有关网页搜索的更多信息，请遵循 MiniMax ChatCompletion 。
About web search follow the MiniMax ChatCompletion for further information.
以下是如何使用网页搜索的简单代码片段：
Here is a simple snippet how to use the web search:
UserMessage userMessage = new UserMessage(
"How many gold medals has the United States won in total at the 2024 Olympics?");
List<Message> messages = new ArrayList<>(List.of(this.userMessage));
List<MiniMaxApi.FunctionTool> functionTool = List.of(MiniMaxApi.FunctionTool.webSearchFunctionTool());
MiniMaxChatOptions options = MiniMaxChatOptions.builder()
.model(MiniMaxApi.ChatModel.ABAB_6_5_S_Chat.value)
.tools(this.functionTool)
.build();
// Sync request
ChatResponse response = chatModel.call(new Prompt(this.messages, this.options));
// Streaming request
Flux<ChatResponse> streamResponse = chatModel.stream(new Prompt(this.messages, this.options));MiniMaxApi Samples
-
MiniMaxApiIT.java 测试提供了一些关于如何使用轻量级库的通用示例。
-
The MiniMaxApiIT.java test provides some general examples how to use the lightweight library.
-
MiniMaxApiToolFunctionCallIT.java 测试展示了如何使用低级 API 调用工具函数。
-
The MiniMaxApiToolFunctionCallIT.java test shows how to use the low-level API to call tool functions.>