ai-world-sdk 1.0.6 → 1.0.8

package/README.md

@@ -29,7 +29,7 @@ npm install ai-world-sdk
 import { sdkConfig } from 'ai-world-sdk';
 
 // 设置全局配置（只需一次）
-sdkConfig.setBaseUrl('http://localhost:8000');
+sdkConfig.setBaseUrl('your-base-url');
 sdkConfig.setToken('your-jwt-token');
 sdkConfig.setDebug(true); // 可选：启用调试模式
 
@@ -98,7 +98,10 @@ console.log('图像 URL:', result.data[0]?.url);
 const geminiClient = new GeminiImageGenerationClient({});
 const geminiResult = await geminiClient.generate({
   prompt: 'A futuristic city',
+  model: 'gemini-3-pro-image-preview', // 使用 Gemini 3 Pro 模型
   aspect_ratio: '16:9',
+  image_size: '2K', // 仅适用于 gemini-3-pro-image-preview
+  response_modalities: ['IMAGE'], // 仅返回图片，不返回文本
 });
 console.log('图像 URL:', geminiResult.data[0]?.url);
 ```
@@ -211,15 +214,68 @@ const result = await client.generate({
 ```typescript
 const client = new GeminiImageGenerationClient({});
 
+// 基础用法（使用 Gemini 2.5 Flash - 快速模型）
 const result = await client.generate({
   prompt: 'A beautiful landscape',              // 必需
-  model: 'gemini-2.0-flash-exp-image-generation', // 可选，默认值
-  number_of_images: 1,                          // 可选
-  aspect_ratio: '16:9',                         // 可选: 1:1, 9:16, 16:9, 4:3, 3:4
-  temperature: 0.7,                             // 可选
+  model: 'gemini-2.5-flash-image',              // 推荐：快速、高效
+  aspect_ratio: '16:9',                         // 可选: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9
+  number_of_images: 1,                          // 可选: 1-4
+});
+
+// 高级用法（使用 Gemini 3 Pro - 专业模型，支持高分辨率）
+const result2 = await client.generate({
+  prompt: 'A futuristic city at sunset',
+  model: 'gemini-3-pro-image-preview',         // 专业级模型
+  aspect_ratio: '16:9',
+  image_size: '2K',                             // 可选: 1K, 2K, 4K（仅适用于 gemini-3-pro-image-preview）
+  number_of_images: 1,
+  response_modalities: ['IMAGE'],              // 仅返回图片，不返回文本
+  temperature: 0.7,                             // 可选: 0.0-2.0
+  max_output_tokens: 1000,                      // 可选
+});
+
+// 使用 aihubmix provider（通过 aihubmix 代理服务）
+const aihubmixClient = new GeminiImageGenerationClient({
+  provider: 'aihubmix',  // 使用 aihubmix 代理
+});
+
+const result3 = await aihubmixClient.generate({
+  prompt: 'A beautiful landscape with mountains',
+  model: 'gemini-3-pro-image-preview',
+  aspect_ratio: '16:9',
+  image_size: '1K',
+  number_of_images: 1,
+  response_modalities: ['IMAGE'],
 });
 ```
 
+**参数说明：**
+
+| 参数 | 类型 | 说明 | 默认值 |
+|------|------|------|--------|
+| `prompt` | `string` | 图像生成提示词（必需） | - |
+| `model` | `string` | 模型名称 | `gemini-2.0-flash-exp-image-generation` |
+| `aspect_ratio` | `string` | 宽高比 | - |
+| `image_size` | `string` | 图片大小（仅 gemini-3-pro-image-preview） | - |
+| `number_of_images` | `number` | 生成图片数量 | `1` |
+| `response_modalities` | `array` | 响应模态 | `['TEXT', 'IMAGE']` |
+| `temperature` | `number` | 温度参数 | `0.7` |
+| `max_output_tokens` | `number` | 最大输出 token 数 | `1000` |
+
+**支持的宽高比：** `1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`
+
+**Provider 说明：**
+
+- `provider: 'gemini'`（默认）- 直接使用 Google Gemini 官方 API
+- `provider: 'aihubmix'` - 通过 aihubmix 代理服务调用，适合需要统一管理多个 API 的场景
+
+**模型对比：**
+
+| 模型 | 分辨率 | 特点 | 适用场景 |
+|------|--------|------|----------|
+| `gemini-2.5-flash-image` | 1024px | 快速、高效、成本低 | 日常使用、批量生成 |
+| `gemini-3-pro-image-preview` | 1K/2K/4K | 专业级、高分辨率、高级功能 | 专业设计、高分辨率需求 |
+
 ### 视频生成
 
 #### VideoGenerationClient
@@ -253,7 +309,7 @@ const result = await client.poll(task.id, {
 import { sdkConfig } from 'ai-world-sdk';
 
 // 设置配置
-sdkConfig.setBaseUrl('http://localhost:8000');
+sdkConfig.setBaseUrl('your-base-url');
 sdkConfig.setToken('your-jwt-token');
 sdkConfig.setHeaders({ 'X-Custom-Header': 'value' });
 sdkConfig.setDebug(true);
@@ -308,11 +364,26 @@ const gptModel = new ChatOpenAI({
 
 // Gemini 模型（通过 aihubmix）
 const geminiModel = new ChatGoogleGenerativeAI({
-  modelName: 'gemini-2.5-flash-image',
+  modelName: 'gemini-2.5-flash',
   temperature: 0.7,
   provider: 'aihubmix',
 });
 
+// Gemini 图像生成（通过 aihubmix）
+import { GeminiImageGenerationClient } from 'ai-world-sdk';
+
+const geminiImageClient = new GeminiImageGenerationClient({
+  provider: 'aihubmix',
+});
+
+const imageResult = await geminiImageClient.generate({
+  prompt: 'A beautiful landscape',
+  model: 'gemini-3-pro-image-preview',
+  aspect_ratio: '16:9',
+  image_size: '1K',
+  response_modalities: ['IMAGE'],
+});
+
 // Claude 模型（通过 aihubmix）
 const claudeModel = new ChatAnthropic({
   modelName: 'claude-3-sonnet-20240229',
@@ -444,6 +515,8 @@ const response = await modelWithTools.invoke([
 
 ### 图像生成工作流
 
+#### 豆包图像生成
+
 ```typescript
 import { DoubaoImageGenerationClient } from 'ai-world-sdk';
 
@@ -470,6 +543,38 @@ multiResult.data.forEach((image, index) => {
 });
 ```
 
+#### Gemini 图像生成
+
+```typescript
+import { GeminiImageGenerationClient } from 'ai-world-sdk';
+
+const client = new GeminiImageGenerationClient({});
+
+// 使用 Gemini 2.5 Flash（快速模型）
+const result = await client.generate({
+  prompt: 'A beautiful sunset over the ocean',
+  model: 'gemini-2.5-flash-image',
+  aspect_ratio: '16:9',
+  number_of_images: 1,
+});
+
+console.log('图像 URL:', result.data[0]?.url || 'Base64 编码');
+if (result.text) {
+  console.log('图像描述:', result.text);
+}
+
+// 使用 Gemini 3 Pro（专业模型，支持高分辨率）
+const result2 = await client.generate({
+  prompt: 'A futuristic city at night',
+  model: 'gemini-3-pro-image-preview',
+  aspect_ratio: '21:9', // 超宽屏
+  image_size: '4K',    // 4K 分辨率
+  response_modalities: ['IMAGE'], // 仅返回图片
+});
+
+console.log('4K 图像:', result2.data[0]?.b64_json ? 'Base64 编码' : result2.data[0]?.url);
+```
+
 ### 视频生成工作流
 
 ```typescript
@@ -519,7 +624,14 @@ if (result.status === 'succeeded') {
 ### 图像生成模型
 
 - **豆包 Seedream**: `doubao-seedream-4-5-251128` (默认)
-- **Google Gemini**: `gemini-2.0-flash-exp-image-generation` (默认)
+- **Google Gemini**:
+  - `gemini-2.5-flash-image` (Nano Banana) - **推荐**，快速、高效，1024px 分辨率，支持所有宽高比
+  - `gemini-3-pro-image-preview` (Nano Banana Pro) - 专业级，支持 1K/2K/4K 分辨率，支持 Google 搜索、思考模式，最多 14 张参考图片
+  - `gemini-2.0-flash-exp-image-generation` (已弃用，建议使用 `gemini-2.5-flash-image`)
+
+**模型选择建议：**
+- **日常使用**: `gemini-2.5-flash-image` - 速度快，成本低
+- **专业需求**: `gemini-3-pro-image-preview` - 高分辨率、高级功能（需要 `image_size` 参数）
 
 ### 视频生成模型
 

package/dist/__tests__/example.test.js

@@ -228,38 +228,6 @@ describe("Langchain SDK Tests", () => {
             console.log("响应内容:", content);
         }
     }, 60000);
-    test("ChatGoogleGenerativeAI - aihubmix.com 图像生成", async () => {
-        const gemini = new index_1.ChatGoogleGenerativeAI({
-            modelName: "gemini-3-pro-image-preview",
-            temperature: 0.7,
-            provider: "aihubmix",
-        });
-        const response = await gemini.invoke([
-            new index_1.HumanMessage("请生成一张展示人工智能概念的图片"),
-        ]);
-        expect(response).toBeDefined();
-        expect(response.content).toBeDefined();
-        // 图像生成响应可能包含文本和图像数据
-        // 检查响应内容（可能是字符串、数组或包含图像的对象）
-        const content = response.content;
-        if (typeof content === "string") {
-            console.log("✅ aihubmix.com 图像生成测试成功（文本响应）");
-            console.log("AI:", content);
-        }
-        else if (Array.isArray(content)) {
-            // 多模态响应，可能包含图像
-            const hasImage = content.some((item) => item?.type === "image_url" ||
-                item?.image_url ||
-                (typeof item === "string" && item.startsWith("data:image")));
-            console.log("✅ aihubmix.com 图像生成测试成功（多模态响应）");
-            console.log("响应类型: 数组，包含图像:", hasImage);
-            console.log("响应内容:", JSON.stringify(content, null, 2));
-        }
-        else {
-            console.log("✅ aihubmix.com 图像生成测试成功（其他格式）");
-            console.log("响应内容:", content);
-        }
-    }, 60000);
     test("DoubaoImageGenerationClient - 基础图像生成", async () => {
         const imageClient = new index_1.DoubaoImageGenerationClient({});
         const result = await imageClient.generate({
@@ -589,10 +557,12 @@ describe("Langchain SDK Tests", () => {
     test("GeminiImageGenerationClient - 基础图像生成", async () => {
         const imageClient = new index_1.GeminiImageGenerationClient({});
         const result = await imageClient.generate({
-            prompt: "A beautiful sunset over the ocean with vibrant colors",
-            model: "gemini-2.0-flash-exp-image-generation",
+            prompt: 'A beautiful sunset over the ocean',
+            model: 'gemini-3-pro-image-preview',
+            aspect_ratio: '16:9',
+            image_size: '1K', // 仅适用于 gemini-3-pro-image-preview
             number_of_images: 1,
-            aspect_ratio: "16:9",
+            response_modalities: ['IMAGE'], // 仅返回图片
         });
         expect(result).toBeDefined();
         expect(result.created).toBeDefined();
@@ -606,46 +576,40 @@ describe("Langchain SDK Tests", () => {
         });
         console.log("✅ GeminiImageGenerationClient 基础测试成功");
         console.log(`生成图像数量: ${result.data.length}`);
-        console.log("图像 URL:", result.data[0]?.url || "Base64 编码");
+        console.log("图像 URL:", result.data[0]?.url || ("Base64 编码" + result.data[0]?.b64_json));
         if (result.text) {
             console.log("图像描述:", result.text);
         }
     }, 120000);
-    test("GeminiImageGenerationClient - 多图像生成", async () => {
-        const imageClient = new index_1.GeminiImageGenerationClient({});
+    test("GeminiImageGenerationClient - 使用 aihubmix provider", async () => {
+        const imageClient = new index_1.GeminiImageGenerationClient({
+            provider: "aihubmix",
+        });
         const result = await imageClient.generate({
-            prompt: "A futuristic city skyline",
-            model: "gemini-2.0-flash-exp-image-generation",
-            number_of_images: 2,
-            aspect_ratio: "16:9",
+            prompt: 'A beautiful sunset over the ocean',
+            model: 'gemini-3-pro-image-preview',
+            aspect_ratio: '16:9',
+            image_size: '1K',
+            number_of_images: 1,
+            response_modalities: ['IMAGE'], // 仅返回图片
         });
         expect(result).toBeDefined();
+        expect(result.created).toBeDefined();
+        expect(typeof result.created).toBe("number");
         expect(result.data).toBeDefined();
         expect(Array.isArray(result.data)).toBe(true);
-        expect(result.data.length).toBe(2);
-        console.log("✅ GeminiImageGenerationClient 多图像生成测试成功");
+        expect(result.data.length).toBeGreaterThan(0);
+        result.data.forEach((item) => {
+            expect(item).toBeDefined();
+            expect(item.url || item.b64_json).toBeDefined();
+        });
+        console.log("✅ GeminiImageGenerationClient (aihubmix provider) 测试成功");
         console.log(`生成图像数量: ${result.data.length}`);
-    }, 120000);
-    test("GeminiImageGenerationClient - 不同宽高比测试", async () => {
-        const imageClient = new index_1.GeminiImageGenerationClient({});
-        const aspectRatios = [
-            "1:1",
-            "16:9",
-            "9:16",
-        ];
-        for (const aspectRatio of aspectRatios) {
-            const result = await imageClient.generate({
-                prompt: "A beautiful landscape",
-                model: "gemini-2.0-flash-exp-image-generation",
-                aspect_ratio: aspectRatio,
-                number_of_images: 1,
-            });
-            expect(result).toBeDefined();
-            expect(result.data).toBeDefined();
-            expect(result.data.length).toBeGreaterThan(0);
-            console.log(`✅ Gemini 宽高比 ${aspectRatio} 测试成功`);
+        console.log("图像 URL:", result.data[0]?.url || ("Base64 编码" + result.data[0]?.b64_json));
+        if (result.text) {
+            console.log("图像描述:", result.text);
         }
-    }, 180000);
+    }, 120000);
     test("DoubaoImageGenerationClient - quality 参数测试", async () => {
         const imageClient = new index_1.DoubaoImageGenerationClient({});
         const qualities = ["standard", "hd"];
@@ -660,7 +624,7 @@ describe("Langchain SDK Tests", () => {
             expect(result).toBeDefined();
             expect(result.data).toBeDefined();
             expect(result.data.length).toBeGreaterThan(0);
-            console.log(`✅ quality ${quality} 测试成功`);
+            console.log(`✅ quality ${quality} 测试成功`, result.data);
         }
     }, 180000);
     test("DoubaoImageGenerationClient - style 参数测试", async () => {
@@ -677,7 +641,7 @@ describe("Langchain SDK Tests", () => {
             expect(result).toBeDefined();
             expect(result.data).toBeDefined();
             expect(result.data.length).toBeGreaterThan(0);
-            console.log(`✅ style ${style} 测试成功`);
+            console.log(`✅ style ${style} 测试成功`, result.data);
         }
     }, 180000);
     test("DoubaoImageGenerationClient - response_format 参数测试", async () => {
@@ -717,7 +681,7 @@ describe("Langchain SDK Tests", () => {
         expect(result).toBeDefined();
         expect(result.data).toBeDefined();
         expect(result.data.length).toBeGreaterThan(0);
-        console.log("✅ watermark 参数测试成功");
+        console.log("✅ watermark 参数测试成功", result.data);
     }, 120000);
     test("sdkConfig - 全局配置测试", () => {
         const originalBaseUrl = index_1.sdkConfig.getServerUrl();

package/dist/gemini-image-generation.d.ts

@@ -3,6 +3,7 @@
  * Google Gemini 图像生成客户端
  */
 export interface GeminiImageGenerationConfig {
+    provider?: "aihubmix" | "gemini";
     baseUrl?: string;
     headers?: Record<string, string>;
 }
@@ -10,9 +11,11 @@ export interface GeminiImageGenerationRequest {
     prompt: string;
     model?: string;
     number_of_images?: number;
-    aspect_ratio?: "1:1" | "9:16" | "16:9" | "4:3" | "3:4";
+    aspect_ratio?: "1:1" | "2:3" | "3:2" | "3:4" | "4:3" | "4:5" | "5:4" | "9:16" | "16:9" | "21:9";
+    image_size?: "1K" | "2K" | "4K";
     temperature?: number;
     max_output_tokens?: number;
+    response_modalities?: ("TEXT" | "IMAGE")[];
     user?: string;
 }
 export interface GeminiImageData {
@@ -27,10 +30,17 @@ export interface GeminiImageGenerationResponse {
 }
 export declare class GeminiImageGenerationClient {
     private headers;
+    private provider;
     constructor(config?: GeminiImageGenerationConfig);
     /**
      * Generate images
      * 生成图像
+     *
+     * 支持的参数：
+     * - aspect_ratio: 宽高比，支持 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9
+     * - image_size: 图片大小（仅 gemini-3-pro-image-preview），支持 1K, 2K, 4K
+     * - response_modalities: 响应模态，["TEXT", "IMAGE"] 或 ["IMAGE"]
+     * - number_of_images: 生成图片数量（1-4）
      */
     generate(request: GeminiImageGenerationRequest): Promise<GeminiImageGenerationResponse>;
 }

package/dist/gemini-image-generation.js

@@ -9,19 +9,31 @@ const config_1 = require("./config");
 const log_1 = require("./log");
 class GeminiImageGenerationClient {
     constructor(config) {
+        // 确定 provider（默认为 gemini）
+        this.provider = config?.provider || "gemini";
         // 合并全局 headers 和配置 headers
         const globalHeaders = config_1.sdkConfig.getHeaders();
         this.headers = {
             "Content-Type": "application/json",
             "Authorization": `Bearer ${config_1.sdkConfig.getToken()}`,
-            "X-Base-Url": config?.baseUrl || "",
+            "X-Provider": this.provider, // 在 header 中传递 provider，由后端确定 baseUrl
             ...globalHeaders,
             ...config?.headers,
         };
+        // 如果明确指定了 baseUrl，也添加到 header 中
+        if (config?.baseUrl) {
+            this.headers["X-Base-Url"] = config.baseUrl;
+        }
     }
     /**
      * Generate images
      * 生成图像
+     *
+     * 支持的参数：
+     * - aspect_ratio: 宽高比，支持 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9
+     * - image_size: 图片大小（仅 gemini-3-pro-image-preview），支持 1K, 2K, 4K
+     * - response_modalities: 响应模态，["TEXT", "IMAGE"] 或 ["IMAGE"]
+     * - number_of_images: 生成图片数量（1-4）
      */
     async generate(request) {
         const requestBody = {
@@ -33,12 +45,18 @@ class GeminiImageGenerationClient {
         if (request.aspect_ratio) {
             requestBody.aspect_ratio = request.aspect_ratio;
         }
+        if (request.image_size) {
+            requestBody.image_size = request.image_size;
+        }
         if (request.temperature !== undefined) {
             requestBody.temperature = request.temperature;
         }
         if (request.max_output_tokens) {
             requestBody.max_output_tokens = request.max_output_tokens;
         }
+        if (request.response_modalities) {
+            requestBody.response_modalities = request.response_modalities;
+        }
         if (request.user) {
             requestBody.user = request.user;
         }

package/dist/index.d.ts

@@ -10,7 +10,7 @@ import { GeminiImageGenerationClient, type GeminiImageGenerationConfig, type Gem
 import { VideoGenerationClient, type VideoGenerationConfig, type VideoGenerationRequest, type ContentGenerationTaskID, type ContentGenerationTask } from "./video_generation";
 import { sdkConfig } from "./config";
 export { BaseMessage, HumanMessage, AIMessage, SystemMessage, AIMessageChunk, type MessageContent, type AIMessageChunkData, } from "./messages";
-export { BaseChatModel, type BaseChatModelParams, type ToolDefinition, type BindOptions, } from "./base";
+export { BaseChatModel, type BaseChatModelParams, type AIModelProvider, type ToolDefinition, type BindOptions, } from "./base";
 export { ChatOpenAI } from "./chat_models/openai";
 export { ChatGoogleGenerativeAI } from "./chat_models/google";
 export { ChatAnthropic } from "./chat_models/anthropic";

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ai-world-sdk",
-  "version": "1.0.6",
+  "version": "1.0.8",
   "description": "TypeScript SDK for AI World Platform - Chat Models, Image Generation, and Video Generation",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",