fix: update test cases (#6)

* fix: update test cases

* feat: add new model constants and update AudioServiceTest to use them

* feat: update model list in README and README_CN with new AI capabilities

---------

Co-authored-by: tomsun28 <tomsun28@outlook.com>

README.md

@@ -329,16 +329,53 @@ The ZaiClient provides access to comprehensive AI services:
 - `glm-4` - Standard GLM-4 model
 - `glm-4-air` - Lightweight version for speed
 - `glm-4-flash` - Ultra-fast response model
+- `glm-4-0520` - GLM-4 model version 0520
+- `glm-4-airx` - Extended Air model with additional features
 - `glm-4-long` - Optimized for long-context conversations
 - `glm-4-voice` - Specialized for voice interactions
+- `glm-4.1v-thinking-flash` - Visual reasoning model with thinking capabilities
+- `glm-z1-air` - Optimized for mathematical and logical reasoning
+- `glm-z1-airx` - Fastest domestic inference model with 200 tokens/s
+- `glm-z1-flash` - Completely free reasoning model service
+- `glm-4-air-250414` - Enhanced with reinforcement learning optimization
+- `glm-4-flash-250414` - Latest free language model
+- `glm-4-flashx` - Enhanced Flash version with ultra-fast inference speed
+- `glm-4-9b` - Open-source model with 9 billion parameters
+- `glm-4-assistant` - AI assistant for various business scenarios
+- `glm-4-alltools` - Agent model for complex task planning and execution
+- `chatglm3-6b` - Open-source base model with 6 billion parameters
+- `codegeex-4` - Code generation and completion model
+
+### Audio Speech Recognition
+- `glm-asr` - Context-aware audio transcription model
+
+### Real-time Interaction
+- `glm-realtime-air` - Real-time video call model with cross-modal reasoning
+- `glm-realtime-flash` - Fast real-time video call model
 
 ### Vision Models
 - `glm-4v-plus` - Enhanced vision model
 - `glm-4v` - Standard vision model
+- `glm-4v-plus-0111` - Variable resolution video and image understanding
+- `glm-4v-flash` - Free and powerful image understanding model
 
 ### Image Generation
 - `cogview-3-plus` - Enhanced image generation
 - `cogview-3` - Standard image generation
+- `cogview-3-flash` - Free image generation model
+- `cogview-4-250304` - Advanced image generation with text capabilities
+- `cogview-4` - Advanced image generation for precise and personalized AI image expression
+
+### Video Generation
+- `cogvideox` - Video generation from text or images
+- `cogvideox-flash` - Free video generation model
+- `cogvideox-2` - New video generation model
+- `viduq1-text` - High-performance video generation from text input
+- `viduq1-image` - Video generation from first frame image and text description
+- `viduq1-start-end` - Video generation from first and last frame images
+- `vidu2-image` - Enhanced video generation from first frame image and text description
+- `vidu2-start-end` - Enhanced video generation from first and last frame images
+- `vidu2-reference` - Video generation with reference images of people, objects, etc.
 
 ### Embeddings
 - `embedding-3` - Latest embedding model
@@ -347,6 +384,7 @@ The ZaiClient provides access to comprehensive AI services:
 ### Specialized
 - `charglm-3` - Character interaction model
 - `cogtts` - Text-to-speech model
+- `rerank` - Text reordering and relevance scoring
 
 ## 📈 Release Notes
 

README_CN.md

@@ -352,28 +352,66 @@ public class AIController {
 ## 🤖 支持的模型
 
 ### 文本生成模型
-- `glm-4-plus` - 最新的GLM-4 Plus模型
-- `glm-4-0520` - GLM-4标准版
-- `glm-4-long` - 长文本处理版本
-- `glm-4-airx` - 轻量级版本
-- `glm-4-air` - 快速响应版本
-- `glm-4-flashx` - 超快响应版本
-- `glm-4-flash` - 闪电版本
+- `glm-4-plus` - 增强版GLM-4，具有更强的能力
+- `glm-4` - 标准GLM-4模型
+- `glm-4-air` - 轻量级版本，优化速度
+- `glm-4-flash` - 超快响应模型
+- `glm-4-0520` - GLM-4模型版本0520
+- `glm-4-airx` - 扩展Air模型，具有附加功能
+- `glm-4-long` - 优化长上下文对话
+- `glm-4-voice` - 专为语音交互设计
+- `glm-4.1v-thinking-flash` - 具有思维能力的视觉推理模型
+- `glm-z1-air` - 优化数学和逻辑推理
+- `glm-z1-airx` - 国内最快推理模型，200 tokens/s
+- `glm-z1-flash` - 完全免费的推理模型服务
+- `glm-4-air-250414` - 通过强化学习优化增强
+- `glm-4-flash-250414` - 最新免费语言模型
+- `glm-4-flashx` - 增强Flash版本，具有超快推理速度
+- `glm-4-9b` - 90亿参数开源模型
+- `glm-4-assistant` - 面向各种业务场景的AI助手
+- `glm-4-alltools` - 复杂任务规划和执行的代理模型
+- `chatglm3-6b` - 60亿参数开源基础模型
+- `codegeex-4` - 代码生成和补全模型
+
+### 音频语音识别
+- `glm-asr` - 上下文感知音频转录模型
+
+### 实时交互
+- `glm-realtime-air` - 具有跨模态推理的实时视频通话模型
+- `glm-realtime-flash` - 快速实时视频通话模型
 
 ### 视觉模型
-- `glm-4v-plus` - 多模态理解模型
-- `glm-4v` - 视觉理解模型
+- `glm-4v-plus` - 增强视觉模型
+- `glm-4v` - 标准视觉模型
+- `glm-4v-plus-0111` - 可变分辨率视频和图像理解
+- `glm-4v-flash` - 免费且强大的图像理解模型
 
 ### 图像生成模型
-- `cogview-3-plus` - 高质量图像生成
+- `cogview-3-plus` - 增强图像生成
 - `cogview-3` - 标准图像生成
+- `cogview-3-flash` - 免费图像生成模型
+- `cogview-4-250304` - 具有文本功能的高级图像生成
+- `cogview-4` - 精确个性化AI图像表达的高级图像生成
+
+### 视频生成模型
+- `cogvideox` - 从文本或图像生成视频
+- `cogvideox-flash` - 免费视频生成模型
+- `cogvideox-2` - 新视频生成模型
+- `viduq1-text` - 从文本输入的高性能视频生成
+- `viduq1-image` - 从首帧图像和文本描述生成视频
+- `viduq1-start-end` - 从首末帧图像生成视频
+- `vidu2-image` - 从首帧图像和文本描述的增强视频生成
+- `vidu2-start-end` - 从首末帧图像的增强视频生成
+- `vidu2-reference` - 使用人物、物体等参考图像的视频生成
 
 ### 嵌入模型
 - `embedding-3` - 最新嵌入模型
-- `embedding-2` - 标准嵌入模型
+- `embedding-2` - 上一代嵌入模型
 
 ### 专业模型
-- `charglm-3` - 角色扮演模型
+- `charglm-3` - 角色交互模型
+- `cogtts` - 文本转语音模型
+- `rerank` - 文本重排序和相关性评分
 - `emohaa` - 情感分析模型
 
 ## 📈 版本更新

core/src/main/java/ai/z/openapi/core/Constants.java

@@ -75,6 +75,93 @@ public final class Constants {
 	 */
 	public static final String ModelChatGLM4Voice = "glm-4-voice";
 
+	/**
+	 * GLM-4.1V Thinking Flash model - Visual reasoning model with thinking capabilities.
+	 */
+	public static final String ModelChatGLM41VThinkingFlash = "glm-4.1v-thinking-flash";
+
+	/**
+	 * GLM-Z1 Air model - Optimized for mathematical and logical reasoning.
+	 */
+	public static final String ModelChatGLMZ1Air = "glm-z1-air";
+
+	/**
+	 * GLM-Z1 AirX model - Fastest domestic inference model with 200 tokens/s.
+	 */
+	public static final String ModelChatGLMZ1AirX = "glm-z1-airx";
+
+	/**
+	 * GLM-Z1 Flash model - Completely free reasoning model service.
+	 */
+	public static final String ModelChatGLMZ1Flash = "glm-z1-flash";
+
+	/**
+	 * GLM-4 Air 250414 model - Enhanced with reinforcement learning optimization.
+	 */
+	public static final String ModelChatGLM4Air250414 = "glm-4-air-250414";
+
+	/**
+	 * GLM-4 Flash 250414 model - Latest free language model.
+	 */
+	public static final String ModelChatGLM4Flash250414 = "glm-4-flash-250414";
+
+	/**
+	 * GLM-4 FlashX model - Enhanced Flash version with ultra-fast inference speed.
+	 */
+	public static final String ModelChatGLM4FlashX = "glm-4-flashx";
+
+	/**
+	 * GLM-4 9B model - Open-source model with 9 billion parameters.
+	 */
+	public static final String ModelChatGLM49B = "glm-4-9b";
+
+	/**
+	 * GLM-4 Assistant model - AI assistant for various business scenarios.
+	 */
+	public static final String ModelChatGLM4Assistant = "glm-4-assistant";
+
+	/**
+	 * GLM-4 AllTools model - Agent model for complex task planning and execution.
+	 */
+	public static final String ModelChatGLM4AllTools = "glm-4-alltools";
+
+	/**
+	 * ChatGLM3 6B model - Open-source base model with 6 billion parameters.
+	 */
+	public static final String ModelChatGLM36B = "chatglm3-6b";
+
+	/**
+	 * CodeGeeX-4 model - Code generation and completion model.
+	 */
+	public static final String ModelCodeGeeX4 = "codegeex-4";
+
+	// =============================================================================
+	// Audio Speech Recognition Models
+	// =============================================================================
+
+	/**
+	 * GLM-ASR model - Context-aware audio transcription model that converts audio to
+	 * fluent and readable text. Supports Chinese, English, and various Chinese dialects.
+	 * Improved performance in noisy environments.
+	 */
+	public static final String ModelGLMASR = "glm-asr";
+
+	// =============================================================================
+	// Real-time Interaction Models
+	// =============================================================================
+
+	/**
+	 * GLM-Realtime Air model - Real-time video call model with cross-modal reasoning
+	 * capabilities across text, audio, and video. Supports real-time interruption.
+	 */
+	public static final String ModelGLMRealtimeAir = "glm-realtime-air";
+
+	/**
+	 * GLM-Realtime Flash model - Fast real-time video call model with cross-modal
+	 * reasoning capabilities. Supports camera interaction and screen sharing.
+	 */
+	public static final String ModelGLMRealtimeFlash = "glm-realtime-flash";
+
 	// =============================================================================
 	// Vision Models (Image Understanding)
 	// =============================================================================
@@ -89,6 +176,16 @@ public final class Constants {
 	 */
 	public static final String ModelChatGLM4V = "glm-4v";
 
+	/**
+	 * GLM-4V Plus 0111 model - Variable resolution video and image understanding.
+	 */
+	public static final String ModelChatGLM4VPlus0111 = "glm-4v-plus-0111";
+
+	/**
+	 * GLM-4V Flash model - Free and powerful image understanding model.
+	 */
+	public static final String ModelChatGLM4VFlash = "glm-4v-flash";
+
 	// =============================================================================
 	// Image Generation Models
 	// =============================================================================
@@ -103,6 +200,75 @@ public final class Constants {
 	 */
 	public static final String ModelCogView = "cogview-3";
 
+	/**
+	 * CogView-3 Flash model - Free image generation model.
+	 */
+	public static final String ModelCogView3Flash = "cogview-3-flash";
+
+	/**
+	 * CogView-4 250304 model - Advanced image generation with text capabilities.
+	 */
+	public static final String ModelCogView4250304 = "cogview-4-250304";
+
+	/**
+	 * CogView-4 model - Advanced image generation for precise and personalized AI image
+	 * expression.
+	 */
+	public static final String ModelCogView4 = "cogview-4";
+
+	// =============================================================================
+	// Video Generation Models
+	// =============================================================================
+
+	/**
+	 * CogVideoX model - Video generation from text or images.
+	 */
+	public static final String ModelCogVideoX = "cogvideox";
+
+	/**
+	 * CogVideoX Flash model - Free video generation model.
+	 */
+	public static final String ModelCogVideoXFlash = "cogvideox-flash";
+
+	/**
+	 * CogVideoX-2 model - New video generation model.
+	 */
+	public static final String ModelCogVideoX2 = "cogvideox-2";
+
+	/**
+	 * Vidu Q1 Text model - High-performance video generation from text input. Supports
+	 * general and anime styles.
+	 */
+	public static final String ModelViduQ1Text = "viduq1-text";
+
+	/**
+	 * Vidu Q1 Image model - Video generation from first frame image and text description.
+	 */
+	public static final String ModelViduQ1Image = "viduq1-image";
+
+	/**
+	 * Vidu Q1 Start-End model - Video generation from first and last frame images.
+	 */
+	public static final String ModelViduQ1StartEnd = "viduq1-start-end";
+
+	/**
+	 * Vidu 2 Image model - Enhanced video generation from first frame image and text
+	 * description.
+	 */
+	public static final String ModelVidu2Image = "vidu2-image";
+
+	/**
+	 * Vidu 2 Start-End model - Enhanced video generation from first and last frame
+	 * images.
+	 */
+	public static final String ModelVidu2StartEnd = "vidu2-start-end";
+
+	/**
+	 * Vidu 2 Reference model - Video generation with reference images of people, objects,
+	 * etc.
+	 */
+	public static final String ModelVidu2Reference = "vidu2-reference";
+
 	// =============================================================================
 	// Embedding Models
 	// =============================================================================
@@ -131,23 +297,9 @@ public final class Constants {
 	 */
 	public static final String ModelTTS = "cogtts";
 
-	// =============================================================================
-	// API Invocation Methods
-	// =============================================================================
-
 	/**
-	 * Asynchronous invocation method - For non-blocking API calls.
+	 * Rerank model - Text reordering and relevance scoring.
 	 */
-	public static final String INVOKE_METHOD_ASYNC = "async-invoke";
-
-	/**
-	 * Server-Sent Events invocation method - For streaming responses.
-	 */
-	public static final String INVOKE_METHOD_SSE = "sse-invoke";
-
-	/**
-	 * Standard synchronous invocation method - For blocking API calls.
-	 */
-	public static final String INVOKE_METHOD = "invoke";
+	public static final String ModelRerank = "rerank";
 
 }

core/src/main/java/ai/z/openapi/service/image/ImageResult.java

@@ -32,11 +32,4 @@ public class ImageResult {
 	@JsonFormat(with = JsonFormat.Feature.ACCEPT_SINGLE_VALUE_AS_ARRAY)
 	List<Image> data;
 
-	/**
-	 * Task number submitted by user in client request or task number generated by
-	 * platform
-	 */
-	@JsonProperty("request_id")
-	private String requestId;
-
 }

core/src/test/java/ai/z/openapi/service/audio/AudioServiceTest.java

@@ -1,6 +1,7 @@
 package ai.z.openapi.service.audio;
 
 import ai.z.openapi.ZaiClient;
+import ai.z.openapi.core.Constants;
 import ai.z.openapi.core.config.ZaiConfig;
 import ai.z.openapi.service.model.ChatCompletionResponse;
 import ai.z.openapi.service.model.Choice;
@@ -62,9 +63,9 @@ public class AudioServiceTest {
 		String requestId = String.format(REQUEST_ID_TEMPLATE, System.currentTimeMillis());
 
 		AudioSpeechRequest request = AudioSpeechRequest.builder()
-			.model("tts-1")
+			.model(Constants.ModelTTS)
 			.input("Hello, this is a test for text-to-speech functionality.")
-			.voice("alloy")
+			.voice("tongtong")
 			.requestId(requestId)
 			.build();
 
@@ -90,7 +91,7 @@ public class AudioServiceTest {
 		File voiceFile = new File("src/test/resources/asr.wav");
 
 		AudioCustomizationRequest request = AudioCustomizationRequest.builder()
-			.model("tts-1")
+			.model(Constants.ModelTTS)
 			.input("This is a test for custom voice generation.")
 			.voiceData(voiceFile)
 			.voiceText("Sample voice text for cloning")
@@ -120,7 +121,7 @@ public class AudioServiceTest {
 		File audioFile = new File("src/test/resources/asr.wav");
 
 		AudioTranscriptionsRequest request = AudioTranscriptionsRequest.builder()
-			.model("glm-asr")
+			.model(Constants.ModelGLMASR)
 			.file(audioFile)
 			.stream(false)
 			.requestId(requestId)
@@ -145,10 +146,10 @@ public class AudioServiceTest {
 	void testStreamAudioTranscription() throws JsonProcessingException {
 		// Prepare test data
 		String requestId = String.format(REQUEST_ID_TEMPLATE, System.currentTimeMillis());
-		File audioFile = new File("src/test/resources/asr.webm");
+		File audioFile = new File("src/test/resources/asr.wav");
 
 		AudioTranscriptionsRequest request = AudioTranscriptionsRequest.builder()
-			.model("glm-asr")
+			.model(Constants.ModelGLMASR)
 			.file(audioFile)
 			.stream(true)
 			.requestId(requestId)
@@ -214,7 +215,7 @@ public class AudioServiceTest {
 	@DisplayName("Test Parameter Validation - Invalid Audio File")
 	void testValidation_InvalidAudioFile() {
 		AudioTranscriptionsRequest request = AudioTranscriptionsRequest.builder()
-			.model("glm-asr")
+			.model(Constants.ModelGLMASR)
 			.file(new File("non-existent-file.wav"))
 			.stream(false)
 			.build();
@@ -227,7 +228,7 @@ public class AudioServiceTest {
 	@Test
 	@DisplayName("Test Parameter Validation - Null Model in Speech Request")
 	void testValidation_NullModelInSpeechRequest() {
-		AudioSpeechRequest request = AudioSpeechRequest.builder().input("Test input").voice("alloy").build();
+		AudioSpeechRequest request = AudioSpeechRequest.builder().input("Test input").voice("tongtong").build();
 
 		assertThrows(IllegalArgumentException.class, () -> {
 			audioService.createSpeech(request);
@@ -237,7 +238,11 @@ public class AudioServiceTest {
 	@Test
 	@DisplayName("Test Parameter Validation - Empty Input in Speech Request")
 	void testValidation_EmptyInputInSpeechRequest() {
-		AudioSpeechRequest request = AudioSpeechRequest.builder().model("tts-1").input("").voice("alloy").build();
+		AudioSpeechRequest request = AudioSpeechRequest.builder()
+			.model(Constants.ModelTTS)
+			.input("")
+			.voice("tongtong")
+			.build();
 
 		assertThrows(IllegalArgumentException.class, () -> {
 			audioService.createSpeech(request);
@@ -248,7 +253,7 @@ public class AudioServiceTest {
 	@DisplayName("Test Parameter Validation - Null Voice Data in Custom Speech Request")
 	void testValidation_NullVoiceDataInCustomSpeechRequest() {
 		AudioCustomizationRequest request = AudioCustomizationRequest.builder()
-			.model("tts-1")
+			.model(Constants.ModelTTS)
 			.input("Test input")
 			.voiceText("Voice text")
 			.build();
@@ -262,7 +267,7 @@ public class AudioServiceTest {
 	@DisplayName("Test Parameter Validation - Null File in Transcription Request")
 	void testValidation_NullFileInTranscriptionRequest() {
 		AudioTranscriptionsRequest request = AudioTranscriptionsRequest.builder()
-			.model("glm-asr")
+			.model(Constants.ModelGLMASR)
 			.stream(false)
 			.build();
 
@@ -275,13 +280,15 @@ public class AudioServiceTest {
 	@DisplayName("Test Speech Generation with Different Voice Options")
 	@EnabledIfEnvironmentVariable(named = "ZAI_API_KEY", matches = "^[^.]+\\.[^.]+$")
 	void testSpeechGenerationWithDifferentVoices() throws JsonProcessingException {
-		String[] voices = { "alloy", "echo", "fable", "onyx", "nova", "shimmer" };
+		// Test with available voice options (currently limited to 'tongtong' as of
+		// 2025-01-14)
+		String[] voices = { "tongtong" };
 
 		for (String voice : voices) {
 			String requestId = String.format(REQUEST_ID_TEMPLATE + "-%s", System.currentTimeMillis(), voice);
 
 			AudioSpeechRequest request = AudioSpeechRequest.builder()
-				.model("tts-1")
+				.model(Constants.ModelTTS)
 				.input("Testing voice: " + voice)
 				.voice(voice)
 				.requestId(requestId)
@@ -305,7 +312,7 @@ public class AudioServiceTest {
 			File file = new File("src/test/resources/" + audioFile);
 
 			AudioTranscriptionsRequest request = AudioTranscriptionsRequest.builder()
-				.model("glm-asr")
+				.model(Constants.ModelGLMASR)
 				.file(file)
 				.stream(false)
 				.requestId(requestId)

core/src/test/java/ai/z/openapi/service/chat/ChatServiceTest.java

@@ -449,7 +449,7 @@ public class ChatServiceTest {
 		stop.add("<|observation|>");
 
 		Map<String, Object> extraJson = new HashMap<>();
-		extraJson.put("invoke_method", Constants.INVOKE_METHOD);
+		extraJson.put("invoke_method", "invoke");
 
 		ChatCompletionCreateParams request = ChatCompletionCreateParams.builder()
 			.model("codegeex-4")

core/src/test/java/ai/z/openapi/service/document/DocumentServiceTest.java

@@ -63,7 +63,7 @@ public class DocumentServiceTest {
 		String requestId = String.format(REQUEST_ID_TEMPLATE, System.currentTimeMillis());
 
 		DocumentCreateParams request = DocumentCreateParams.builder()
-			.filePath("/path/to/test/document.pdf")
+			.filePath("src/test/resources/document.pdf")
 			.purpose("retrieval")
 			.knowledgeId(TEST_KNOWLEDGE_ID)
 			.sentenceSize(200)
@@ -193,7 +193,7 @@ public class DocumentServiceTest {
 		uploadDetail.setUrl("https://example.com/test.pdf");
 
 		DocumentCreateParams request = DocumentCreateParams.builder()
-			.filePath("/path/to/file.pdf")
+			.filePath("src/test/resources/document.pdf\"")
 			.uploadDetail(Collections.singletonList(uploadDetail))
 			.purpose("retrieval")
 			.knowledgeId(TEST_KNOWLEDGE_ID)
@@ -208,7 +208,7 @@ public class DocumentServiceTest {
 	@DisplayName("Test Parameter Validation - Empty Knowledge ID")
 	void testValidation_EmptyKnowledgeId() {
 		DocumentCreateParams request = DocumentCreateParams.builder()
-			.filePath("/path/to/file.pdf")
+			.filePath("src/test/resources/document.pdf")
 			.purpose("retrieval")
 			.knowledgeId("")
 			.build();

core/src/test/java/ai/z/openapi/service/image/ImageServiceTest.java

@@ -72,7 +72,6 @@ public class ImageServiceTest {
 		assertNotNull(response, "Response should not be null");
 		assertTrue(response.isSuccess(), "Response should be successful");
 		assertNotNull(response.getData(), "Response data should not be null");
-		assertEquals(requestId, response.getData().getRequestId(), "Request ID should match");
 		assertNotNull(response.getData().getData(), "Image data should not be null");
 		assertFalse(response.getData().getData().isEmpty(), "Image data should not be empty");
 		assertNull(response.getError(), "Response error should be null");

core/src/test/java/ai/z/openapi/service/videos/VideosServiceTest.java

@@ -71,7 +71,7 @@ public class VideosServiceTest {
 			.prompt("A beautiful sunset over the ocean with waves gently crashing on the shore")
 			.requestId(requestId)
 			.withAudio(Boolean.TRUE)
-			.quality("standard")
+			.quality("speed")
 			.duration(5)
 			.build();