EasyFlow-Tech-Dev/Easy-Agents
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

初始化

Browse Source
This commit is contained in:
陈子默
2026-02-22 18:55:40 +08:00
commit 8392cdd861
496 changed files with 45020 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
easy-agents-chat/easy-agents-chat-deepseek/pom.xml Normal file
View File

@@ -0,0 +1,30 @@
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-chat</artifactId>
<version>${revision}</version>
</parent>
<packaging>jar</packaging>
<name>easy-agents-chat-deepseek</name>
<artifactId>easy-agents-chat-deepseek</artifactId>
<properties>
<maven.compiler.source>8</maven.compiler.source>
<maven.compiler.target>8</maven.compiler.target>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>
<dependencies>
<dependency>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-core</artifactId>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<scope>test</scope>
</dependency>
</dependencies>
</project>

52
easy-agents-chat/easy-agents-chat-deepseek/src/main/java/com/easyagents/llm/deepseek/DeepseekChatModel.java Normal file
View File

@@ -0,0 +1,52 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.deepseek;
import com.easyagents.core.model.chat.OpenAICompatibleChatModel;
import com.easyagents.core.model.chat.ChatInterceptor;
import com.easyagents.core.model.chat.GlobalChatInterceptors;
import java.util.List;
/**
* @author huangjf
* @version : v1.0
*/
public class DeepseekChatModel extends OpenAICompatibleChatModel<DeepseekConfig> {
/**
* 构造一个聊天模型实例，不使用实例级拦截器。
*
* @param config 聊天模型配置
*/
public DeepseekChatModel(DeepseekConfig config) {
super(config);
}
/**
* 构造一个聊天模型实例，并指定实例级拦截器。
* <p>
* 实例级拦截器会与全局拦截器（通过 {@link GlobalChatInterceptors} 注册）合并，
* 执行顺序为：可观测性拦截器 → 全局拦截器 → 实例拦截器。
*
* @param config 聊天模型配置
* @param userInterceptors 实例级拦截器列表
*/
public DeepseekChatModel(DeepseekConfig config, List<ChatInterceptor> userInterceptors) {
super(config, userInterceptors);
}
}

36
easy-agents-chat/easy-agents-chat-deepseek/src/main/java/com/easyagents/llm/deepseek/DeepseekConfig.java Normal file
View File

@@ -0,0 +1,36 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.deepseek;
import com.easyagents.core.model.chat.ChatConfig;
/**
* @author huangjf
* @version : v1.0
*/
public class DeepseekConfig extends ChatConfig {
private static final String DEFAULT_MODEL = "deepseek-chat";
private static final String DEFAULT_ENDPOINT = "https://api.deepseek.com";
private static final String DEFAULT_REQUEST_PATH = "/chat/completions";
public DeepseekConfig() {
setEndpoint(DEFAULT_ENDPOINT);
setRequestPath(DEFAULT_REQUEST_PATH);
setModel(DEFAULT_MODEL);
}
}

92
easy-agents-chat/easy-agents-chat-deepseek/src/test/java/com/easyagents/llm/deepseek/DeepseekTest.java Normal file
View File

@@ -0,0 +1,92 @@
package com.easyagents.llm.deepseek;
import com.easyagents.core.message.UserMessage;
import com.easyagents.core.model.client.StreamContext;
import com.easyagents.core.model.chat.ChatModel;
import com.easyagents.core.model.chat.StreamResponseListener;
import com.easyagents.core.model.chat.tool.annotation.ToolDef;
import com.easyagents.core.model.chat.tool.annotation.ToolParam;
import com.easyagents.core.model.chat.response.AiMessageResponse;
import com.easyagents.core.message.SystemMessage;
import com.easyagents.core.prompt.MemoryPrompt;
import com.easyagents.core.prompt.SimplePrompt;
import com.easyagents.core.util.StringUtil;
import java.util.Scanner;
public class DeepseekTest {
@ToolDef(name = "get_the_weather_info", description = "get the weather info")
public static String getWeatherInfo(@ToolParam(name = "city", description = "城市名称") String name) {
//在这里，我们应该通过第三方接口调用 api 信息
return name + "的天气是阴转多云。 ";
}
@ToolDef(name = "get_holiday_balance", description = "获取假期余额")
public static String getHolidayBalance() {
//在这里，我们应该通过第三方接口调用 api 信息
String username = "michael";
return username + "你的年假还剩余3天，有效期至26年1月。调休假剩余1天，长期有效。 ";
}
public static ChatModel getLLM() {
DeepseekConfig deepseekConfig = new DeepseekConfig();
deepseekConfig.setEndpoint("https://api.siliconflow.cn/v1");
deepseekConfig.setApiKey("*********************");
deepseekConfig.setModel("Pro/deepseek-ai/DeepSeek-V3");
deepseekConfig.setLogEnabled(true);
return new DeepseekChatModel(deepseekConfig);
}
public static void chatHr() {
ChatModel chatModel = getLLM();
MemoryPrompt prompt = new MemoryPrompt();
// 加入system
prompt.addMessage(new SystemMessage("你是一个人事助手小智，专注于为用户提供高效、精准的信息查询和问题解答服务。"));
System.out.println("我是小智，你的人事小助手！请尽情吩咐小智！");
Scanner scanner = new Scanner(System.in);
String userInput = scanner.nextLine();
while (userInput != null) {
// 第二步：创建 HumanMessage，并添加方法调用
UserMessage userMessage = new UserMessage(userInput);
userMessage.addToolsFromClass(DeepseekTest.class);
// 第三步：将 HumanMessage 添加到 HistoriesPrompt 中
prompt.addMessage(userMessage);
// 第四步：调用 chatStream 方法，进行对话
chatModel.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
if (StringUtil.hasText(response.getMessage().getContent())) {
System.out.print(response.getMessage().getContent());
}
if (response.getMessage().isFinalDelta()) {
System.out.println(response);
System.out.println("------");
}
}
@Override
public void onStop(StreamContext context) {
System.out.println("stop!!!------");
}
});
userInput = scanner.nextLine();
}
}
public static void functionCall() {
ChatModel chatModel = getLLM();
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(DeepseekTest.class);
AiMessageResponse response = chatModel.chat(prompt);
System.out.println(response.executeToolCallsAndGetResults());
}
public static void main(String[] args) {
// functionCall();
chatHr();
}
}

34
easy-agents-chat/easy-agents-chat-ollama/pom.xml Normal file
View File

@@ -0,0 +1,34 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-chat</artifactId>
<version>${revision}</version>
</parent>
<name>easy-agents-chat-ollama</name>
<artifactId>easy-agents-chat-ollama</artifactId>
<properties>
<maven.compiler.source>8</maven.compiler.source>
<maven.compiler.target>8</maven.compiler.target>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>
<dependencies>
<dependency>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-core</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<scope>test</scope>
</dependency>
</dependencies>
</project>

32
easy-agents-chat/easy-agents-chat-ollama/src/main/java/com/easyagents/llm/ollama/OllamaChatConfig.java Normal file
View File

@@ -0,0 +1,32 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.ollama;
import com.easyagents.core.model.chat.ChatConfig;
public class OllamaChatConfig extends ChatConfig {
private static final String DEFAULT_PROVIDER = "ollama";
private static final String DEFAULT_ENDPOINT = "https://localhost:11434";
private static final String DEFAULT_REQUEST_PATH = "/v1/chat/completions";
public OllamaChatConfig() {
setProvider(DEFAULT_PROVIDER);
setEndpoint(DEFAULT_ENDPOINT);
setRequestPath(DEFAULT_REQUEST_PATH);
}
}

56
easy-agents-chat/easy-agents-chat-ollama/src/main/java/com/easyagents/llm/ollama/OllamaChatModel.java Normal file
View File

@@ -0,0 +1,56 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.ollama;
import com.easyagents.core.model.chat.OpenAICompatibleChatModel;
import com.easyagents.core.model.chat.ChatInterceptor;
import com.easyagents.core.model.chat.GlobalChatInterceptors;
import com.easyagents.core.model.client.ChatRequestSpecBuilder;
import java.util.List;
public class OllamaChatModel extends OpenAICompatibleChatModel<OllamaChatConfig> {
/**
* 构造一个聊天模型实例，不使用实例级拦截器。
*
* @param config 聊天模型配置
*/
public OllamaChatModel(OllamaChatConfig config) {
super(config);
}
/**
* 构造一个聊天模型实例，并指定实例级拦截器。
* <p>
* 实例级拦截器会与全局拦截器（通过 {@link GlobalChatInterceptors} 注册）合并，
* 执行顺序为：可观测性拦截器 → 全局拦截器 → 实例拦截器。
*
* @param config 聊天模型配置
* @param userInterceptors 实例级拦截器列表
*/
public OllamaChatModel(OllamaChatConfig config, List<ChatInterceptor> userInterceptors) {
super(config, userInterceptors);
}
@Override
public ChatRequestSpecBuilder getChatRequestSpecBuilder() {
return new OllamaRequestSpecBuilder();
}
}

21
easy-agents-chat/easy-agents-chat-ollama/src/main/java/com/easyagents/llm/ollama/OllamaRequestSpecBuilder.java Normal file
View File

@@ -0,0 +1,21 @@
package com.easyagents.llm.ollama;
import com.easyagents.core.model.chat.ChatConfig;
import com.easyagents.core.model.chat.ChatOptions;
import com.easyagents.core.model.client.OpenAIChatRequestSpecBuilder;
import com.easyagents.core.prompt.Prompt;
import com.easyagents.core.util.Maps;
public class OllamaRequestSpecBuilder extends OpenAIChatRequestSpecBuilder {
protected Maps buildBaseParamsOfRequestBody(Prompt prompt, ChatOptions options, ChatConfig config) {
Maps params = super.buildBaseParamsOfRequestBody(prompt, options, config);
params.setIf(!options.isStreaming(), "stream", false);
// 支持思考
if (config.isSupportThinking()) {
params.setIf(options.getThinkingEnabled() != null, "thinking", options.getThinkingEnabled());
}
return params;
}
}

94
easy-agents-chat/easy-agents-chat-ollama/src/test/java/com/easyagents/llm/ollama/OllamaChatModelTest.java Normal file
View File

@@ -0,0 +1,94 @@
package com.easyagents.llm.ollama;
import com.easyagents.core.message.AiMessage;
import com.easyagents.core.model.chat.ChatModel;
import com.easyagents.core.model.chat.response.AiMessageResponse;
import com.easyagents.core.model.exception.ModelException;
import com.easyagents.core.prompt.SimplePrompt;
import org.junit.Test;
public class OllamaChatModelTest {
@Test(expected = ModelException.class)
public void testChat() {
OllamaChatConfig config = new OllamaChatConfig();
config.setEndpoint("http://localhost:11434");
config.setModel("llama3");
config.setLogEnabled(true);
ChatModel chatModel = new OllamaChatModel(config);
String chat = chatModel.chat("Why is the sky blue?");
System.out.println(">>>" + chat);
}
@Test
public void testChatStream() throws InterruptedException {
OllamaChatConfig config = new OllamaChatConfig();
config.setEndpoint("http://localhost:11434");
config.setModel("llama3");
config.setLogEnabled(true);
ChatModel chatModel = new OllamaChatModel(config);
chatModel.chatStream("Why is the sky blue?", (context, response) -> System.out.println(response.getMessage().getContent()));
Thread.sleep(2000);
}
@Test
public void testFunctionCall1() throws InterruptedException {
OllamaChatConfig config = new OllamaChatConfig();
config.setEndpoint("http://localhost:11434");
config.setModel("llama3.1");
config.setLogEnabled(true);
ChatModel chatModel = new OllamaChatModel(config);
SimplePrompt prompt = new SimplePrompt("What's the weather like in Beijing?");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = chatModel.chat(prompt);
System.out.println(response.executeToolCallsAndGetResults());
}
@Test
public void testFunctionCall2() throws InterruptedException {
OllamaChatConfig config = new OllamaChatConfig();
config.setEndpoint("http://localhost:11434");
config.setModel("llama3.1");
config.setLogEnabled(true);
ChatModel chatModel = new OllamaChatModel(config);
SimplePrompt prompt = new SimplePrompt("What's the weather like in Beijing?");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = chatModel.chat(prompt);
if (response.hasToolCalls()) {
prompt.setToolMessages(response.executeToolCallsAndGetToolMessages());
AiMessageResponse response1 = chatModel.chat(prompt);
System.out.println(response1.getMessage().getContent());
}
}
@Test
public void testVisionModel() {
OllamaChatConfig config = new OllamaChatConfig();
config.setEndpoint("http://localhost:11434");
config.setModel("llava");
config.setLogEnabled(true);
ChatModel chatModel = new OllamaChatModel(config);
SimplePrompt imagePrompt = new SimplePrompt("What's in the picture?");
imagePrompt.addImageUrl("https://agentsflex.com/assets/images/logo.png");
AiMessageResponse response = chatModel.chat(imagePrompt);
AiMessage message = response == null ? null : response.getMessage();
System.out.println(message);
}
}

22
easy-agents-chat/easy-agents-chat-ollama/src/test/java/com/easyagents/llm/ollama/WeatherFunctions.java Normal file
View File

@@ -0,0 +1,22 @@
package com.easyagents.llm.ollama;
import com.easyagents.core.model.chat.tool.annotation.ToolDef;
import com.easyagents.core.model.chat.tool.annotation.ToolParam;
public class WeatherFunctions {
@ToolDef(name = "get_the_weather_info", description = "get the weather info")
public static String getWeatherInfo(
@ToolParam(name = "city", description = "the city name") String name
) {
return "Snowy days";
}
@ToolDef(name = "get_the_temperature", description = "get the temperature")
public static String getTemperature(
@ToolParam(name = "city", description = "the city name") String name
) {
return "The temperature in " + name + " is 15°C";
}
}

32
easy-agents-chat/easy-agents-chat-openai/pom.xml Normal file
View File

@@ -0,0 +1,32 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-chat</artifactId>
<version>${revision}</version>
</parent>
<name>easy-agents-chat-openai</name>
<artifactId>easy-agents-chat-openai</artifactId>
<properties>
<maven.compiler.source>8</maven.compiler.source>
<maven.compiler.target>8</maven.compiler.target>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>
<dependencies>
<dependency>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-core</artifactId>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<scope>test</scope>
</dependency>
</dependencies>
</project>

214
easy-agents-chat/easy-agents-chat-openai/src/main/java/com/easyagents/llm/openai/OpenAIChatConfig.java Normal file
View File

@@ -0,0 +1,214 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.openai;
import com.easyagents.core.model.chat.ChatConfig;
import com.easyagents.core.model.chat.ChatInterceptor;
import com.easyagents.core.util.StringUtil;
import java.util.List;
import java.util.Map;
/**
* OpenAI 聊天模型的配置类，支持通过 Builder 模式创建配置或直接构建 {@link OpenAIChatModel}。
* <p>
* 默认值：
* <ul>
* <li>provider: {@code "openai"}</li>
* <li>model: {@code "gpt-3.5-turbo"}</li>
* <li>endpoint: {@code "https://api.openai.com"}</li>
* <li>requestPath: {@code "/v1/chat/completions"}</li>
* </ul>
* <p>
* 该配置类专为 OpenAI 兼容 API 设计，适用于 OpenAI 官方、Azure OpenAI 或其他兼容服务。
*/
public class OpenAIChatConfig extends ChatConfig {
private static final String DEFAULT_PROVIDER = "openai";
private static final String DEFAULT_MODEL = "gpt-3.5-turbo";
private static final String DEFAULT_ENDPOINT = "https://api.openai.com";
private static final String DEFAULT_REQUEST_PATH = "/v1/chat/completions";
public OpenAIChatConfig() {
setProvider(DEFAULT_PROVIDER);
setEndpoint(DEFAULT_ENDPOINT);
setRequestPath(DEFAULT_REQUEST_PATH);
setModel(DEFAULT_MODEL);
}
/**
* 创建一个 {@link OpenAIChatModel} 实例，使用当前配置。
*
* @return 新的 {@link OpenAIChatModel} 实例
*/
public final OpenAIChatModel toChatModel() {
return new OpenAIChatModel(this);
}
/**
* 创建一个 {@link OpenAIChatModel} 实例，使用当前配置和指定的实例级拦截器。
*
* @param interceptors 实例级拦截器列表，可为 {@code null} 或空列表
* @return 新的 {@link OpenAIChatModel} 实例
*/
public final OpenAIChatModel toChatModel(List<ChatInterceptor> interceptors) {
return new OpenAIChatModel(this, interceptors);
}
/**
* 构建器类，用于流畅地创建 {@link OpenAIChatConfig} 或直接构建 {@link OpenAIChatModel}。
*/
public static class Builder {
private final OpenAIChatConfig config = new OpenAIChatConfig();
// --- BaseModelConfig fields ---
public Builder apiKey(String apiKey) {
config.setApiKey(apiKey);
return this;
}
public Builder provider(String provider) {
config.setProvider(provider);
return this;
}
public Builder endpoint(String endpoint) {
config.setEndpoint(endpoint);
return this;
}
public Builder requestPath(String requestPath) {
config.setRequestPath(requestPath);
return this;
}
public Builder model(String model) {
config.setModel(model);
return this;
}
/**
* 添加单个自定义属性（会进行深拷贝，不会持有外部引用）。
*/
public Builder customProperty(String key, Object value) {
config.putCustomProperty(key, value);
return this;
}
/**
* 设置自定义属性映射（会进行深拷贝，不会持有外部 map 引用）。
*/
public Builder customProperties(Map<String, Object> customProperties) {
config.setCustomProperties(customProperties);
return this;
}
// --- ChatConfig fields ---
public Builder supportImage(Boolean supportImage) {
config.setSupportImage(supportImage);
return this;
}
public Builder supportImageBase64Only(Boolean supportImageBase64Only) {
config.setSupportImageBase64Only(supportImageBase64Only);
return this;
}
public Builder supportAudio(Boolean supportAudio) {
config.setSupportAudio(supportAudio);
return this;
}
public Builder supportVideo(Boolean supportVideo) {
config.setSupportVideo(supportVideo);
return this;
}
public Builder supportTool(Boolean supportTool) {
config.setSupportTool(supportTool);
return this;
}
public Builder supportThinking(Boolean supportThinking) {
config.setSupportThinking(supportThinking);
return this;
}
public Builder thinkingEnabled(boolean thinkingEnabled) {
config.setThinkingEnabled(thinkingEnabled);
return this;
}
public Builder observabilityEnabled(boolean observabilityEnabled) {
config.setObservabilityEnabled(observabilityEnabled);
return this;
}
public Builder logEnabled(boolean logEnabled) {
config.setLogEnabled(logEnabled);
return this;
}
/**
* 构建 {@link OpenAIChatConfig} 配置对象。
* <p>
* 该方法会校验必要字段（如 {@code apiKey}），若缺失将抛出异常。
*
* @return 构建完成的配置对象
* @throws IllegalStateException 如果 {@code apiKey} 未设置或为空
*/
public OpenAIChatConfig build() {
if (StringUtil.noText(config.getApiKey())) {
throw new IllegalStateException("apiKey must be set for OpenAIChatConfig");
}
return config;
}
/**
* 直接构建 {@link OpenAIChatModel} 实例，使用默认（全局）拦截器。
*
* @return 新的聊天模型实例
* @throws IllegalStateException 如果 {@code apiKey} 未设置或为空
*/
public OpenAIChatModel buildModel() {
return new OpenAIChatModel(build());
}
/**
* 直接构建 {@link OpenAIChatModel} 实例，并指定实例级拦截器。
*
* @param interceptors 实例级拦截器列表，可为 {@code null} 或空
* @return 新的聊天模型实例
* @throws IllegalStateException 如果 {@code apiKey} 未设置或为空
*/
public OpenAIChatModel buildModel(List<ChatInterceptor> interceptors) {
return new OpenAIChatModel(build(), interceptors);
}
}
/**
* 获取一个新的构建器实例，用于链式配置。
*
* @return {@link Builder} 实例
*/
public static Builder builder() {
return new Builder();
}
}

62
easy-agents-chat/easy-agents-chat-openai/src/main/java/com/easyagents/llm/openai/OpenAIChatModel.java Normal file
View File

@@ -0,0 +1,62 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.openai;
import com.easyagents.core.model.chat.BaseChatModel;
import com.easyagents.core.model.chat.OpenAICompatibleChatModel;
import com.easyagents.core.model.chat.ChatInterceptor;
import com.easyagents.core.model.chat.GlobalChatInterceptors;
import java.util.List;
/**
* OpenAI 聊天模型实现。
* <p>
* 该类封装了 OpenAI API 的具体调用细节，包括：
* <ul>
* <li>请求体构建（支持同步/流式）</li>
* <li>HTTP 客户端管理</li>
* <li>解析器配置（同步/流式使用不同解析器）</li>
* </ul>
* <p>
* 所有横切逻辑（监控、日志、拦截）由 {@link BaseChatModel} 的责任链处理，
* 本类只关注 OpenAI 协议特有的实现细节。
*/
public class OpenAIChatModel extends OpenAICompatibleChatModel<OpenAIChatConfig> {
/**
* 构造一个聊天模型实例，不使用实例级拦截器。
*
* @param config 聊天模型配置
*/
public OpenAIChatModel(OpenAIChatConfig config) {
super(config);
}
/**
* 构造一个聊天模型实例，并指定实例级拦截器。
* <p>
* 实例级拦截器会与全局拦截器（通过 {@link GlobalChatInterceptors} 注册）合并，
* 执行顺序为：可观测性拦截器 → 全局拦截器 → 实例拦截器。
*
* @param config 聊天模型配置
* @param userInterceptors 实例级拦截器列表
*/
public OpenAIChatModel(OpenAIChatConfig config, List<ChatInterceptor> userInterceptors) {
super(config, userInterceptors);
}
}

86
easy-agents-chat/easy-agents-chat-openai/src/test/java/com/easyagents/llm/openai/ChatModelTestUtils.java Normal file
View File

@@ -0,0 +1,86 @@
package com.easyagents.llm.openai;
import com.easyagents.core.model.chat.ChatModel;
import com.easyagents.core.model.chat.ChatOptions;
import com.easyagents.core.model.chat.StreamResponseListener;
import com.easyagents.core.model.chat.response.AiMessageResponse;
import com.easyagents.core.model.client.StreamContext;
import com.easyagents.core.prompt.Prompt;
import com.easyagents.core.prompt.SimplePrompt;
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.TimeUnit;
public class ChatModelTestUtils {
public static void waitForStream(
ChatModel model,
String prompt,
StreamResponseListener listener) {
waitForStream(model, new SimplePrompt(prompt), listener, Integer.MAX_VALUE, null);
}
public static void waitForStream(
ChatModel model,
String prompt,
StreamResponseListener listener,
ChatOptions options) {
waitForStream(model, new SimplePrompt(prompt), listener, Integer.MAX_VALUE, options);
}
public static void waitForStream(
ChatModel model,
Prompt prompt,
StreamResponseListener listener) {
waitForStream(model, prompt, listener, Integer.MAX_VALUE, null);
}
public static void waitForStream(
ChatModel model,
Prompt prompt,
StreamResponseListener listener,
ChatOptions options) {
waitForStream(model, prompt, listener, Integer.MAX_VALUE, options);
}
public static void waitForStream(
ChatModel model,
Prompt prompt,
StreamResponseListener listener,
long timeoutSeconds, ChatOptions options) {
CountDownLatch latch = new CountDownLatch(1);
StreamResponseListener wrapped = new StreamResponseListener() {
@Override
public void onStart(StreamContext context) {
listener.onStart(context);
}
@Override
public void onMessage(StreamContext ctx, AiMessageResponse resp) {
listener.onMessage(ctx, resp);
}
@Override
public void onStop(StreamContext ctx) {
listener.onStop(ctx);
latch.countDown();
}
@Override
public void onFailure(StreamContext context, Throwable throwable) {
listener.onFailure(context, throwable);
}
};
model.chatStream(prompt, wrapped, options);
try {
if (!latch.await(timeoutSeconds, TimeUnit.SECONDS)) {
throw new RuntimeException("Stream did not complete within " + timeoutSeconds + "s");
}
} catch (InterruptedException e) {
throw new RuntimeException(e);
}
}
}

53
easy-agents-chat/easy-agents-chat-openai/src/test/java/com/easyagents/llm/openai/GiteeAiImageTest.java Normal file
View File

@@ -0,0 +1,53 @@
package com.easyagents.llm.openai;
import com.easyagents.core.model.chat.ChatModel;
import com.easyagents.core.model.chat.response.AiMessageResponse;
import com.easyagents.core.prompt.SimplePrompt;
import org.jetbrains.annotations.NotNull;
import org.junit.Test;
public class GiteeAiImageTest {
@NotNull
private static OpenAIChatConfig getOpenAIChatConfig() {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setApiKey("PXW1GXE******L7D12");
// config.setModel("InternVL3-78B");
config.setModel("Qwen3-32B");
config.setEndpoint("https://ai.gitee.com");
config.setLogEnabled(true);
return config;
}
@Test
public void testImage() {
OpenAIChatConfig config = getOpenAIChatConfig();
ChatModel chatModel = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("请识别并输入 markdown，请用中文输出");
prompt.addImageUrl("http://www.codeformat.cn/static/images/logo.png");
AiMessageResponse response = chatModel.chat(prompt);
if (!response.isError()) {
System.out.println(response.getMessage().getContent());
}
}
@Test
public void testChat() {
OpenAIChatConfig config = getOpenAIChatConfig();
config.setSupportImage(false);
ChatModel chatModel = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("你叫什么名字");
prompt.addImageUrl("http://www.codeformat.cn/static/images/logo.png");
AiMessageResponse response = chatModel.chat(prompt);
if (!response.isError()) {
System.out.println(response.getMessage().getContent());
}
}
}

500
easy-agents-chat/easy-agents-chat-openai/src/test/java/com/easyagents/llm/openai/OpenAIChatModelTest.java Normal file
View File

@@ -0,0 +1,500 @@
package com.easyagents.llm.openai;
import com.easyagents.core.agent.react.ReActAgent;
import com.easyagents.core.agent.react.ReActAgentListener;
import com.easyagents.core.agent.react.ReActAgentState;
import com.easyagents.core.agent.react.ReActStep;
import com.easyagents.core.memory.ChatMemory;
import com.easyagents.core.message.ToolCall;
import com.easyagents.core.message.ToolMessage;
import com.easyagents.core.message.UserMessage;
import com.easyagents.core.model.chat.ChatModel;
import com.easyagents.core.model.chat.ChatOptions;
import com.easyagents.core.model.chat.StreamResponseListener;
import com.easyagents.core.model.chat.response.AiMessageResponse;
import com.easyagents.core.model.chat.tool.Tool;
import com.easyagents.core.model.chat.tool.ToolScanner;
import com.easyagents.core.model.client.StreamContext;
import com.easyagents.core.model.exception.ModelException;
import com.easyagents.core.prompt.SimplePrompt;
import org.junit.Test;
import java.util.List;
import java.util.concurrent.TimeUnit;
public class OpenAIChatModelTest {
@Test(expected = ModelException.class)
public void testChat() {
String output = OpenAIChatConfig.builder()
.endpoint("https://ai.gitee.com")
.provider("GiteeAI")
.model("Qwen3-32B")
.apiKey("PXW1****D12")
.buildModel()
.chat("你叫什么名字");
System.out.println(output);
}
@Test()
public void testChatStream() {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setApiKey("PXW1GXE***");
config.setEndpoint("https://ai.gitee.com");
config.setModel("Qwen3-32B");
config.setLogEnabled(true);
ChatOptions options = ChatOptions.builder().thinkingEnabled(false).build();
ChatModel chatModel = new OpenAIChatModel(config);
ChatModelTestUtils.waitForStream(chatModel, "你叫什么名字", new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
System.out.println(response.getMessage().getContent());
}
@Override
public void onFailure(StreamContext context, Throwable throwable) {
System.out.println("onFailure>>>>" + throwable);
}
@Override
public void onStop(StreamContext context) {
System.out.println("stop!!!!");
}
}, options);
}
@Test()
public void testChatStreamBailian() {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setApiKey("sk-32ab******57502");
config.setEndpoint("https://dashscope.aliyuncs.com");
config.setRequestPath("/compatible-mode/v1/chat/completions");
config.setModel("qwen3-max");
ChatModel chatModel = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("北京的天气如何？");
prompt.addToolsFromClass(WeatherFunctions.class);
ChatModelTestUtils.waitForStream(chatModel, prompt, new StreamResponseListener() {
@Override
public void onFailure(StreamContext context, Throwable throwable) {
System.out.println("onFailure>>>>" + throwable);
}
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
if (response.getMessage().getContent() == null) {
System.out.println(response.getMessage());
}
if (response.getMessage().isFinalDelta()) {
List<ToolCall> toolCalls = response.getMessage().getToolCalls();
System.out.println(toolCalls);
}
System.out.println("onMessage >>>>>" + response.getMessage().getContent());
}
@Override
public void onStop(StreamContext context) {
System.out.println("stop!!!!");
}
});
}
@Test
public void testChatOllama() {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setEndpoint("http://localhost:11434");
config.setModel("llama3");
config.setLogEnabled(true);
ChatModel chatModel = new OpenAIChatModel(config);
chatModel.chatStream("who are you", new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
System.out.println(response.getMessage().getContent());
}
@Override
public void onStop(StreamContext context) {
System.out.println("stop!!!!");
}
});
// try {
// Thread.sleep(2000);
// } catch (InterruptedException e) {
// throw new RuntimeException(e);
// }
}
@Test()
public void testChatWithImage() {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setApiKey("sk-5gqOcl*****");
config.setModel("gpt-4-turbo");
ChatModel chatModel = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("What's in this image?");
prompt.addImageUrl("https://upload.wikimedia.org/wikipedia/commons/thumb/d/dd/Gfp-wisconsin-madison-the-nature-boardwalk.jpg/2560px-Gfp-wisconsin-madison-the-nature-boardwalk.jpg");
AiMessageResponse response = chatModel.chat(prompt);
System.out.println(response);
}
@Test()
public void testFunctionCalling1() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setApiKey("sk-rts5NF6n*******");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = llm.chat(prompt);
System.out.println(response.executeToolCallsAndGetResults());
// 阴转多云
}
@Test()
public void testFunctionCalling2() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setApiKey("sk-rts5NF6n*******");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = llm.chat(prompt);
if (response.hasToolCalls()) {
prompt.setToolMessages(response.executeToolCallsAndGetToolMessages());
AiMessageResponse response1 = llm.chat(prompt);
System.out.println(response1.getMessage().getContent());
} else {
System.out.println(response);
}
}
@Test()
public void testFunctionCalling3() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ark.cn-beijing.volces.com");
config.setRequestPath("/api/v3/chat/completions");
config.setModel("deepseek-v3-250324");
config.setApiKey("2d57a");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = llm.chat(prompt);
if (response.hasToolCalls()) {
prompt.setToolMessages(response.executeToolCallsAndGetToolMessages());
AiMessageResponse response1 = llm.chat(prompt);
System.out.println(response1.getMessage().getContent());
} else {
System.out.println(response);
}
}
@Test()
public void testFunctionCalling4() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ark.cn-beijing.volces.com");
config.setRequestPath("/api/v3/chat/completions");
config.setModel("deepseek-v3-250324");
config.setApiKey("2d57aa75");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
System.out.println(" onMessage >>>>>" + response.hasToolCalls());
}
});
TimeUnit.SECONDS.sleep(5);
}
@Test()
public void testFunctionCalling44() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ark.cn-beijing.volces.com");
config.setRequestPath("/api/v3/chat/completions");
config.setModel("deepseek-v3-250324");
config.setApiKey("2d5");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
System.out.println(" onMessage >>>>>" + response.hasToolCalls());
}
});
TimeUnit.SECONDS.sleep(5);
}
@Test()
public void testFunctionCalling444() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ai.gitee.com");
// config.setRequestPath("/api/v3/chat/completions");
// config.setModel("Qwen3-32B");
config.setModel("DeepSeek-V3.2");
config.setApiKey("PXW1G***L7D12");
// config.setLogEnabled(false);
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("北京和上海的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
// System.out.println("onMessage11 >>>>>" + response);
if (response.getMessage().isFinalDelta() && response.hasToolCalls()) {
System.out.println(":::::::: start....");
List<ToolMessage> toolMessages = response.executeToolCallsAndGetToolMessages();
prompt.setAiMessage(response.getMessage());
prompt.setToolMessages(toolMessages);
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
String msg = response.getMessage().getContent() != null ? response.getMessage().getContent() : response.getMessage().getReasoningContent();
System.out.println(":::22" + msg);
}
@Override
public void onStop(StreamContext context) {
System.out.println("onStop >>>>>");
}
});
} else {
String msg = response.getMessage().getContent() != null ? response.getMessage().getContent() : response.getMessage().getReasoningContent();
System.out.println(">>>" + msg);
}
}
});
TimeUnit.SECONDS.sleep(25);
}
@Test()
public void testFunctionCalling5() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ai.gitee.com");
config.setModel("Qwen3-32B");
config.setApiKey("PXW1G*********D12");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("北京和上海的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
// System.out.println("onMessage >>>>>" + response);
}
});
TimeUnit.SECONDS.sleep(25);
}
@Test()
public void testFunctionCalling55() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ai.gitee.com");
config.setModel("Qwen3-32B");
// config.setModel("DeepSeek-V3");
// config.setSupportToolMessage(false);
config.setApiKey("PXW1");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("/no_think 北京和上海的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
if (response.getMessage().isFinalDelta() && response.hasToolCalls()) {
System.out.println(":::::::: start....");
prompt.setAiMessage(response.getMessage());
prompt.setToolMessages(response.executeToolCallsAndGetToolMessages());
llm.chatStream(prompt, new StreamResponseListener() {
@Override
public void onMessage(StreamContext context, AiMessageResponse response) {
String msg = response.getMessage().getContent() != null ? response.getMessage().getContent() : response.getMessage().getReasoningContent();
System.out.println(":::" + msg);
}
});
} else {
String msg = response.getMessage().getContent() != null ? response.getMessage().getContent() : response.getMessage().getReasoningContent();
System.out.println(">>>" + msg);
}
}
});
TimeUnit.SECONDS.sleep(25);
}
@Test()
public void testFunctionCalling6() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
config.setLogEnabled(true);
config.setEndpoint("https://ai.gitee.com");
config.setModel("Qwen3-32B");
config.setApiKey("PXW1");
OpenAIChatModel llm = new OpenAIChatModel(config);
SimplePrompt prompt = new SimplePrompt("/no_think 北京和上海的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = llm.chat(prompt);
prompt.setToolMessages(response.executeToolCallsAndGetToolMessages());
System.out.println(llm.chat(prompt));
}
@Test()
public void testReAct1() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
// config.setDebug(true);
config.setEndpoint("https://ai.gitee.com");
config.setModel("Qwen3-32B");
config.setApiKey("****");
OpenAIChatModel llm = new OpenAIChatModel(config);
List<Tool> tools = ToolScanner.scan(WeatherFunctions.class);
// ReActAgent reActAgent = new ReActAgent(llm, functions, "北京和上海的天气怎么样？");
ReActAgent reActAgent = new ReActAgent(llm, tools, "介绍一下北京");
reActAgent.addListener(new ReActAgentListener() {
@Override
public void onActionStart(ReActStep step) {
System.out.println(">>>>>>" + step.getThought());
System.out.println("正在调用工具 >>>>> " + step.getAction() + ":" + step.getActionInput());
}
@Override
public void onActionEnd(ReActStep step, Object result) {
System.out.println("工具调用结束 >>>>> " + step.getAction() + ":" + step.getActionInput() + ">>>>结果：" + result);
}
@Override
public void onFinalAnswer(String finalAnswer) {
System.out.println("onFinalAnswer >>>>>" + finalAnswer);
}
@Override
public void onNonActionResponse(AiMessageResponse response) {
System.out.println("onNonActionResponse >>>>>" + response.getMessage().getContent());
}
});
reActAgent.execute();
}
@Test()
public void testReAct2() throws InterruptedException {
OpenAIChatConfig config = new OpenAIChatConfig();
// config.setDebug(true);
config.setEndpoint("https://ai.gitee.com");
config.setModel("Qwen2-72B-Instruct");
config.setApiKey("*****");
OpenAIChatModel llm = new OpenAIChatModel(config);
List<Tool> tools = ToolScanner.scan(WeatherFunctions.class);
ReActAgent reActAgent = new ReActAgent(llm, tools, "今天的天气怎么样？");
// reActAgent.setStreamable(true);
reActAgent.addListener(new ReActAgentListener() {
@Override
public void onChatResponseStream(StreamContext context, AiMessageResponse response) {
// System.out.print(response.getMessage().getContent());
}
@Override
public void onRequestUserInput(String question) {
System.out.println("onRequestUserInput>>>" + question);
ReActAgentState state = reActAgent.getState();
state.addMessage(new UserMessage("我在北京市"));
ReActAgent newAgent = new ReActAgent(llm, tools, state);
newAgent.addListener(this);
newAgent.execute();
}
@Override
public void onActionStart(ReActStep step) {
System.out.println(">>>>>>" + step.getThought());
System.out.println("正在调用工具 >>>>> " + step.getAction() + ":" + step.getActionInput());
}
@Override
public void onActionEnd(ReActStep step, Object result) {
System.out.println("工具调用结束 >>>>> " + step.getAction() + ":" + step.getActionInput() + ">>>>结果：" + result);
}
@Override
public void onFinalAnswer(String finalAnswer) {
System.out.println("onFinalAnswer >>>>>" + finalAnswer);
ChatMemory memory = reActAgent.getMemoryPrompt().getMemory();
System.out.println(memory);
}
@Override
public void onNonActionResponseStream(StreamContext context) {
System.out.println("onNonActionResponseStream >>>>>" + context);
}
});
reActAgent.execute();
TimeUnit.SECONDS.sleep(20);
}
}

23
easy-agents-chat/easy-agents-chat-openai/src/test/java/com/easyagents/llm/openai/WeatherFunctions.java Normal file
View File

@@ -0,0 +1,23 @@
package com.easyagents.llm.openai;
import com.easyagents.core.model.chat.tool.annotation.ToolDef;
import com.easyagents.core.model.chat.tool.annotation.ToolParam;
import java.util.concurrent.ThreadLocalRandom;
public class WeatherFunctions {
private static final String[] weathers = {
"晴", "多云", "阴", "小雨", "中雨", "大雨", "暴雨", "雷阵雨",
"小雪", "中雪", "大雪", "暴雪", "雨夹雪", "雾", "霾", "沙尘暴",
"冰雹", "阵雨", "冻雨", "晴间多云", "局部多云", "强对流"
};
@ToolDef(name = "get_the_weather_info", description = "get the weather info")
public static String getWeatherInfo(@ToolParam(name = "city", description = "the city name") String name) {
String weather = weathers[ThreadLocalRandom.current().nextInt(weathers.length)];
System.out.println(">>>>>>>>>>>>>>!!!!!!" + name + ":" + weather);
return weather;
}
}

33
easy-agents-chat/easy-agents-chat-qwen/pom.xml Normal file
View File

@@ -0,0 +1,33 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-chat</artifactId>
<version>${revision}</version>
</parent>
<name>easy-agents-chat-qwen</name>
<artifactId>easy-agents-chat-qwen</artifactId>
<properties>
<maven.compiler.source>8</maven.compiler.source>
<maven.compiler.target>8</maven.compiler.target>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>
<dependencies>
<dependency>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-core</artifactId>
<scope>compile</scope>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<scope>test</scope>
</dependency>
</dependencies>
</project>

32
easy-agents-chat/easy-agents-chat-qwen/src/main/java/com/easyagents/llm/qwen/QwenChatConfig.java Normal file
View File

@@ -0,0 +1,32 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.qwen;
import com.easyagents.core.model.chat.ChatConfig;
public class QwenChatConfig extends ChatConfig {
private static final String DEFAULT_MODEL = "qwen-turbo";
private static final String DEFAULT_ENDPOINT = "https://dashscope.aliyuncs.com";
private static final String DEFAULT_REQUEST_PATH = "/compatible-mode/v1/chat/completions";
public QwenChatConfig() {
setEndpoint(DEFAULT_ENDPOINT);
setRequestPath(DEFAULT_REQUEST_PATH);
setModel(DEFAULT_MODEL);
}
}

55
easy-agents-chat/easy-agents-chat-qwen/src/main/java/com/easyagents/llm/qwen/QwenChatModel.java Normal file
View File

@@ -0,0 +1,55 @@
/*
* Copyright (c) 2023-2026, Easy-Agents (fuhai999@gmail.com).
* <p>
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
* <p>
* http://www.apache.org/licenses/LICENSE-2.0
* <p>
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.easyagents.llm.qwen;
import com.easyagents.core.model.chat.OpenAICompatibleChatModel;
import com.easyagents.core.model.chat.ChatInterceptor;
import com.easyagents.core.model.chat.GlobalChatInterceptors;
import com.easyagents.core.model.client.ChatRequestSpecBuilder;
import java.util.List;
public class QwenChatModel extends OpenAICompatibleChatModel<QwenChatConfig> {
/**
* 构造一个聊天模型实例，不使用实例级拦截器。
*
* @param config 聊天模型配置
*/
public QwenChatModel(QwenChatConfig config) {
super(config);
}
/**
* 构造一个聊天模型实例，并指定实例级拦截器。
* <p>
* 实例级拦截器会与全局拦截器（通过 {@link GlobalChatInterceptors} 注册）合并，
* 执行顺序为：可观测性拦截器 → 全局拦截器 → 实例拦截器。
*
* @param config 聊天模型配置
* @param userInterceptors 实例级拦截器列表
*/
public QwenChatModel(QwenChatConfig config, List<ChatInterceptor> userInterceptors) {
super(config, userInterceptors);
}
@Override
public ChatRequestSpecBuilder getChatRequestSpecBuilder() {
return new QwenRequestSpecBuilder();
}
}

303
easy-agents-chat/easy-agents-chat-qwen/src/main/java/com/easyagents/llm/qwen/QwenChatOptions.java Normal file
View File

@@ -0,0 +1,303 @@
package com.easyagents.llm.qwen;
import com.easyagents.core.model.chat.ChatOptions;
import com.alibaba.fastjson2.annotation.JSONField;
import java.util.List;
/**
* <link href="https://help.aliyun.com/zh/model-studio/use-qwen-by-calling-api">通义千问API参考</link>
*
* @author liutf
*/
public class QwenChatOptions extends ChatOptions {
/**
* 输出数据的模态，仅支持 Qwen-Omni 模型指定。（可选）
* 默认值为["text"]
* 可选值：
* ["text"]：输出文本。
*/
private List<String> modalities;
/**
* 控制模型生成文本时的内容重复度。（可选）
* 取值范围：[-2.0, 2.0]。正数会减少重复度，负数会增加重复度。
* <pre>
* 适用场景：
* 较高的presence_penalty适用于要求多样性、趣味性或创造性的场景，如创意写作或头脑风暴。
* 较低的presence_penalty适用于要求一致性或专业术语的场景，如技术文档或其他正式文档。
* </pre>
* 不建议修改QVQ模型的默认presence_penalty值。
*/
private Float presencePenalty;
/**
* 生成响应的个数，取值范围是1-4。
* 对于需要生成多个响应的场景（如创意写作、广告文案等），可以设置较大的 n 值。
* <pre>
* 当前仅支持 qwen-plus 模型，且在传入 tools 参数时固定为1。
* 设置较大的 n 值不会增加输入 Token 消耗，会增加输出 Token 的消耗。
* </pre>
*/
private Integer n;
/**
* 是否开启并行工具调用。
* 参数为true时开启，为false时不开启。
* 并行工具调用请参见：https://help.aliyun.com/zh/model-studio/qwen-function-calling#cb6b5c484bt4x
*/
private Boolean parallelToolCalls;
/**
* 当您使用翻译模型时需要配置的翻译参数。
*/
private TranslationOptions translationOptions;
/**
* 用于控制模型在生成文本时是否使用互联网搜索结果进行参考。
* 取值如下：
* True：启用互联网搜索，模型会将搜索结果作为文本生成过程中的参考信息，但模型会基于其内部逻辑判断是否使用互联网搜索结果。
* False（默认）：关闭互联网搜索。
* 启用互联网搜索功能可能会增加 Token 的消耗。
* 当前支持 qwen-max、qwen-plus、qwen-turbo
*/
private Boolean enableSearch;
/**
* 思考模式预算，适用于 Qwen3 模型。
* 思考过程的最大长度，只在 enable_thinking 为 true 时生效。适用于 Qwen3 的商业版与开源版模型。
* 详情请参见限制思考长度：https://help.aliyun.com/zh/model-studio/deep-thinking#e7c0002fe4meu
*/
private Integer thinkingBudget;
/**
* 联网搜索的策略。
* 仅当enable_search为true时生效。
*/
private SearchOptions searchOptions;
public static class TranslationOptions {
/**
* （必选）
* 源语言的英文全称
* 您可以将source_lang设置为"auto"，模型会自动判断输入文本属于哪种语言。
* 支持的语言: https://help.aliyun.com/zh/model-studio/user-guide/machine-translation#038d2865bbydc
*/
@JSONField(name = "source_lang")
private String sourceLang;
/**
* （必选）
* 目标语言的英文全称，
* 支持的语言: https://help.aliyun.com/zh/model-studio/user-guide/machine-translation#038d2865bbydc
*/
@JSONField(name = "target_lang")
private String targetLang;
/**
* 在使用术语干预翻译功能时需要设置的术语数组。
* https://help.aliyun.com/zh/model-studio/user-guide/machine-translation#2bf54a5ab5voe
*/
@JSONField(name = "terms")
private List<TranslationOptionsExt> terms;
/**
* 在使用翻译记忆功能时需要设置的翻译记忆数组。
* https://help.aliyun.com/zh/model-studio/user-guide/machine-translation#17e15234e7gfp
*/
@JSONField(name = "tm_list")
private List<TranslationOptionsExt> tmList;
/**
* （可选）
* 在使用领域提示功能时需要设置的领域提示语句。
* 领域提示语句暂时只支持英文。
* https://help.aliyun.com/zh/model-studio/user-guide/machine-translation#4af23a31db7lf
*/
@JSONField(name = "domains")
private String domains;
public String getSourceLang() {
return sourceLang;
}
public TranslationOptions setSourceLang(String sourceLang) {
this.sourceLang = sourceLang;
return this;
}
public String getTargetLang() {
return targetLang;
}
public TranslationOptions setTargetLang(String targetLang) {
this.targetLang = targetLang;
return this;
}
public List<TranslationOptionsExt> getTerms() {
return terms;
}
public TranslationOptions setTerms(List<TranslationOptionsExt> terms) {
this.terms = terms;
return this;
}
public List<TranslationOptionsExt> getTmList() {
return tmList;
}
public TranslationOptions setTmList(List<TranslationOptionsExt> tmList) {
this.tmList = tmList;
return this;
}
public String getDomains() {
return domains;
}
public TranslationOptions setDomains(String domains) {
this.domains = domains;
return this;
}
}
public static class TranslationOptionsExt {
/**
* 源语言的术语/源语言的语句
*/
private String source;
/**
* 目标语言的术语/目标语言的语句
*/
private String target;
public String getSource() {
return source;
}
public TranslationOptionsExt setSource(String source) {
this.source = source;
return this;
}
public String getTarget() {
return target;
}
public TranslationOptionsExt setTarget(String target) {
this.target = target;
return this;
}
}
public static class SearchOptions {
/**
* 是否强制开启搜索。（可选）默认值为false
* 参数值：true=强制开启；false=不强制开启。
*/
@JSONField(name = "forced_search")
private Boolean forcedSearch;
/**
* 搜索互联网信息的数量，（可选）默认值为"standard"
* 参数值：
* standard：在请求时搜索5条互联网信息；
* pro：在请求时搜索10条互联网信息。
*/
@JSONField(name = "search_strategy")
private String searchStrategy;
public Boolean getForcedSearch() {
return forcedSearch;
}
public SearchOptions setForcedSearch(Boolean forcedSearch) {
this.forcedSearch = forcedSearch;
return this;
}
public String getSearchStrategy() {
return searchStrategy;
}
public SearchOptions setSearchStrategy(String searchStrategy) {
this.searchStrategy = searchStrategy;
return this;
}
}
public List<String> getModalities() {
return modalities;
}
public QwenChatOptions setModalities(List<String> modalities) {
this.modalities = modalities;
return this;
}
public Float getPresencePenalty() {
return presencePenalty;
}
public QwenChatOptions setPresencePenalty(Float presencePenalty) {
this.presencePenalty = presencePenalty;
return this;
}
public Integer getN() {
return n;
}
public QwenChatOptions setN(Integer n) {
this.n = n;
return this;
}
public Boolean getParallelToolCalls() {
return parallelToolCalls;
}
public QwenChatOptions setParallelToolCalls(Boolean parallelToolCalls) {
this.parallelToolCalls = parallelToolCalls;
return this;
}
public TranslationOptions getTranslationOptions() {
return translationOptions;
}
public QwenChatOptions setTranslationOptions(TranslationOptions translationOptions) {
this.translationOptions = translationOptions;
return this;
}
public Boolean getEnableSearch() {
return enableSearch;
}
public QwenChatOptions setEnableSearch(Boolean enableSearch) {
this.enableSearch = enableSearch;
return this;
}
public Integer getThinkingBudget() {
return thinkingBudget;
}
public void setThinkingBudget(Integer thinkingBudget) {
this.thinkingBudget = thinkingBudget;
}
public SearchOptions getSearchOptions() {
return searchOptions;
}
public QwenChatOptions setSearchOptions(SearchOptions searchOptions) {
this.searchOptions = searchOptions;
return this;
}
}

30
easy-agents-chat/easy-agents-chat-qwen/src/main/java/com/easyagents/llm/qwen/QwenRequestSpecBuilder.java Normal file
View File

@@ -0,0 +1,30 @@
package com.easyagents.llm.qwen;
import com.easyagents.core.model.chat.ChatConfig;
import com.easyagents.core.model.chat.ChatOptions;
import com.easyagents.core.model.client.OpenAIChatRequestSpecBuilder;
import com.easyagents.core.prompt.Prompt;
import com.easyagents.core.util.CollectionUtil;
import com.easyagents.core.util.Maps;
public class QwenRequestSpecBuilder extends OpenAIChatRequestSpecBuilder {
@Override
protected Maps buildBaseParamsOfRequestBody(Prompt prompt, ChatOptions options, ChatConfig config) {
Maps params = super.buildBaseParamsOfRequestBody(prompt, options, config);
if (options instanceof QwenChatOptions) {
QwenChatOptions op = (QwenChatOptions) options;
params.setIf(CollectionUtil.hasItems(op.getModalities()), "modalities", op.getModalities());
params.setIf(op.getPresencePenalty() != null, "presence_penalty", op.getPresencePenalty());
params.setIf(op.getResponseFormat() != null, "response_format", op.getResponseFormat());
params.setIf(op.getN() != null, "n", op.getN());
params.setIf(op.getParallelToolCalls() != null, "parallel_tool_calls", op.getParallelToolCalls());
params.setIf(op.getTranslationOptions() != null, "translation_options", op.getTranslationOptions());
params.setIf(op.getEnableSearch() != null, "enable_search", op.getEnableSearch());
params.setIf(op.getEnableSearch() != null && op.getEnableSearch() && op.getSearchOptions() != null, "search_options", op.getSearchOptions());
params.setIf(op.getThinkingEnabled() != null, "enable_thinking", op.getThinkingEnabled());
params.setIf(op.getThinkingEnabled() != null && op.getThinkingEnabled() && op.getThinkingBudget() != null, "thinking_budget", op.getThinkingBudget());
}
return params;
}
}

113
easy-agents-chat/easy-agents-chat-qwen/src/test/java/com/easyagents/llm/qwen/test/QwenTest.java Normal file
View File

@@ -0,0 +1,113 @@
package com.easyagents.llm.qwen.test;
import com.easyagents.core.message.AiMessage;
import com.easyagents.core.model.chat.ChatModel;
import com.easyagents.core.model.chat.ChatOptions;
import com.easyagents.core.model.chat.response.AiMessageResponse;
import com.easyagents.core.model.exception.ModelException;
import com.easyagents.core.prompt.SimplePrompt;
import com.easyagents.llm.qwen.QwenChatConfig;
import com.easyagents.llm.qwen.QwenChatModel;
import com.easyagents.llm.qwen.QwenChatOptions;
import com.easyagents.llm.qwen.QwenChatOptions.SearchOptions;
import org.junit.Test;
public class QwenTest {
public static void main(String[] args) throws InterruptedException {
QwenChatConfig config = new QwenChatConfig();
//https://bailian.console.aliyun.com/?apiKey=1#/api-key
config.setApiKey("sk-28a6be3236****");
config.setModel("qwen-plus");
ChatModel chatModel = new QwenChatModel(config);
chatModel.chatStream("请写一个小兔子战胜大灰狼的故事", (context, response) -> {
AiMessage message = response.getMessage();
System.out.println(">>>> " + message.getContent());
});
Thread.sleep(10000);
}
@Test(expected = ModelException.class)
public void testForcedSearch() throws InterruptedException {
QwenChatConfig config = new QwenChatConfig();
config.setApiKey("sk-28a6be3236****");
config.setModel("qwen-max");
ChatModel chatModel = new QwenChatModel(config);
QwenChatOptions options = new QwenChatOptions();
options.setEnableSearch(true);
options.setSearchOptions(new SearchOptions().setForcedSearch(true));
String responseStr = chatModel.chat("今天是几号？", options);
System.out.println(responseStr);
}
@Test
public void testFunctionCalling() throws InterruptedException {
QwenChatConfig config = new QwenChatConfig();
config.setApiKey("sk-28a6be3236****");
config.setModel("qwen-turbo");
ChatModel chatModel = new QwenChatModel(config);
SimplePrompt prompt = new SimplePrompt("今天北京的天气怎么样");
prompt.addToolsFromClass(WeatherFunctions.class);
AiMessageResponse response = chatModel.chat(prompt);
System.out.println(response.executeToolCallsAndGetResults());
// "Today it will be dull and overcast in 北京"
}
/**
* 动态替换模型
*/
@Test
public void testDynamicModel() throws InterruptedException {
// 默认模型
QwenChatConfig config = new QwenChatConfig();
config.setApiKey("sk-28a6be3236****");
config.setModel("qwen-turbo");
// 运行时动态替换模型
ChatOptions options = new QwenChatOptions();
options.setModel("deepseek-r1");
ChatModel chatModel = new QwenChatModel(config);
chatModel.chatStream("请写一个小兔子战胜大灰狼的故事", (context, response) -> {
AiMessage message = response.getMessage();
System.err.println(message.getReasoningContent());
System.out.println(message.getFullContent());
System.out.println();
}, options);
Thread.sleep(10000);
}
/**
* 测试千问3 开启思考模式的开关
*/
@Test
public void testQwen3Thinking() throws InterruptedException {
QwenChatConfig config = new QwenChatConfig();
config.setApiKey("sk-28a6be3236****");
config.setModel("qwen3-235b-a22b");
ChatModel chatModel = new QwenChatModel(config);
QwenChatOptions options = new QwenChatOptions();
options.setThinkingEnabled(false);
//options.setThinkingBudget(1024);
chatModel.chatStream("你是谁", (context, response) -> {
AiMessage message = response.getMessage();
System.err.println(message.getReasoningContent());
System.out.println(message.getFullContent());
System.out.println();
}, options);
Thread.sleep(10000);
}
}

14
easy-agents-chat/easy-agents-chat-qwen/src/test/java/com/easyagents/llm/qwen/test/WeatherFunctions.java Normal file
View File

@@ -0,0 +1,14 @@
package com.easyagents.llm.qwen.test;
import com.easyagents.core.model.chat.tool.annotation.ToolDef;
import com.easyagents.core.model.chat.tool.annotation.ToolParam;
public class WeatherFunctions {
@ToolDef(name = "get_the_weather_info", description = "get the weather info")
public static String getWeatherInfo(
@ToolParam(name = "city", description = "the city name") String name
) {
return "Today it will be dull and overcast in " + name;
}
}

29
easy-agents-chat/pom.xml Normal file
View File

@@ -0,0 +1,29 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>com.easyagents</groupId>
<artifactId>easy-agents-parent</artifactId>
<version>${revision}</version>
</parent>
<name>easy-agents-chat</name>
<artifactId>easy-agents-chat</artifactId>
<packaging>pom</packaging>
<modules>
<module>easy-agents-chat-openai</module>
<module>easy-agents-chat-qwen</module>
<module>easy-agents-chat-ollama</module>
<module>easy-agents-chat-deepseek</module>
</modules>
<properties>
<maven.compiler.source>8</maven.compiler.source>
<maven.compiler.target>8</maven.compiler.target>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
</properties>
</project>