ai-world-sdk 1.0.9 → 1.0.10

package/README.md

@@ -88,7 +88,7 @@ import { DoubaoImageGenerationClient, GeminiImageGenerationClient } from 'ai-wor
 const doubaoClient = new DoubaoImageGenerationClient({});
 const result = await doubaoClient.generate({
   prompt: 'A beautiful sunset over the ocean',
-  size: '2K',
+  size: '2K', // 或使用 '2K', '4K' 等
   quality: 'hd',
   n: 1,
 });
@@ -200,7 +200,7 @@ const client = new DoubaoImageGenerationClient({});
 const result = await client.generate({
   prompt: 'A beautiful landscape',      // 必需
   model: 'doubao-seedream-4-5-251128', // 可选，默认值
-  size: '2K',                          // 可选: 1K, 2K, 4K, 1024x1024 等
+  size: '2K',                   // 可选: 像素值(2048x2048, 2560x1440等)或K值(1K, 2K, 4K)
   quality: 'hd',                       // 可选: standard, hd
   n: 1,                                // 可选: 1-10
   response_format: 'url',              // 可选: url, b64_json
@@ -556,6 +556,27 @@ const response = await modelWithTools.invoke([
 
 #### 豆包图像生成
 
+豆包 Seedream 支持多种图像生成模式：
+
+**支持的尺寸选项：**
+
+- **像素值（宽x高）**：
+  - `2048x2048` (1:1 正方形)
+  - `2304x1728` (4:3 横屏)
+  - `1728x2304` (3:4 竖屏)
+  - `2560x1440` (16:9 横屏)
+  - `1440x2560` (9:16 竖屏)
+  - `2496x1664` (3:2 横屏)
+  - `1664x2496` (2:3 竖屏)
+  - `3024x1296` (21:9 超宽屏)
+
+- **K 值（根据模型版本）**：
+  - `1K` - 仅 4.0 版本支持
+  - `2K` - 4.0 和 4.5 版本支持
+  - `4K` - 4.0 和 4.5 版本支持
+
+**1. 文生图（文本生成图像）**
+
 ```typescript
 import { DoubaoImageGenerationClient } from 'ai-world-sdk';
 
@@ -564,17 +585,58 @@ const client = new DoubaoImageGenerationClient({});
 // 生成单张图像
 const result = await client.generate({
   prompt: 'A futuristic city skyline at sunset',
-  size: '2K',
+  size: '2048x2048', // 1:1 正方形，或使用 '2K', '4K' 等
   quality: 'hd',
 });
 
 console.log('图像 URL:', result.data[0]?.url);
+```
+
+**2. 图文生图（单图输入单图输出）**
+
+基于一张参考图片生成新图像：
+
+```typescript
+// 使用单张图片作为输入
+const result = await client.generate({
+  prompt: '将这张图片转换为水彩画风格',
+  image: 'data:image/png;base64,iVBORw0KGgo...', // base64 编码的图片或 data URL
+  size: '2560x1440', // 16:9 横屏
+  quality: 'hd',
+});
+
+console.log('生成的图像 URL:', result.data[0]?.url);
+```
+
+**3. 多图融合（多图输入单图输出）**
+
+融合多张参考图片生成新图像：
 
-// 生成多张图像
+```typescript
+// 使用多张图片作为输入
+const result = await client.generate({
+  prompt: '将图1的服装换为图2的服装风格',
+  image: [
+    'data:image/png;base64,iVBORw0KGgo...', // 第一张图片
+    'data:image/png;base64,iVBORw0KGgo...', // 第二张图片
+  ],
+  size: '2048x2048', // 1:1 正方形
+  quality: 'hd',
+});
+
+console.log('融合后的图像 URL:', result.data[0]?.url);
+```
+
+**4. 组图输出（多图输出）**
+
+一次生成多张不同风格的图像：
+
+```typescript
+// 生成多张图像（组图输出）
 const multiResult = await client.generate({
   prompt: 'A beautiful landscape',
-  n: 3,
-  size: '2K',
+  n: 3, // 生成 3 张图像
+  size: '2560x1440', // 16:9 横屏
 });
 
 multiResult.data.forEach((image, index) => {
@@ -582,6 +644,19 @@ multiResult.data.forEach((image, index) => {
 });
 ```
 
+**组合使用示例**
+
+```typescript
+// 图文生图 + 组图输出：基于一张图片生成多张不同风格的图像
+const result = await client.generate({
+  prompt: '生成不同风格的艺术作品',
+  image: 'data:image/png;base64,iVBORw0KGgo...',
+  n: 4, // 生成 4 张不同风格的图像
+  size: '2048x2048', // 1:1 正方形
+  quality: 'hd',
+});
+```
+
 #### Gemini 图像生成
 
 ```typescript
@@ -662,7 +737,9 @@ if (result.status === 'succeeded') {
 
 ### 图像生成模型
 
-- **豆包 Seedream**: `doubao-seedream-4-5-251128` (默认)
+- **豆包 Seedream**: 
+  - `doubao-seedream-4-5-251128` (4.5版本，默认) - 支持 2K、4K 和像素值尺寸
+  - `doubao-seedream-4-0` (4.0版本) - 支持 1K、2K、4K 和像素值尺寸
 - **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

package/dist/__tests__/example.test.js

@@ -267,7 +267,7 @@ describe("Langchain SDK Tests", () => {
         expect(result.data).toBeDefined();
         expect(Array.isArray(result.data)).toBe(true);
         expect(result.data.length).toBe(2);
-        console.log("✅ 多图像生成测试成功");
+        console.log("✅ 多图像生成测试成功（组图输出）");
         console.log(`生成图像数量: ${result.data.length}`);
         result.data.forEach((item, index) => {
             console.log(`图像 ${index + 1}:`, item.url || "Base64 编码");
@@ -275,7 +275,10 @@ describe("Langchain SDK Tests", () => {
     }, 120000);
     test("DoubaoImageGenerationClient - 不同尺寸测试", async () => {
         const imageClient = new index_1.DoubaoImageGenerationClient({});
-        const sizes = ["1K", "2K", "4K"];
+        const sizes = [
+            "2K",
+            "4K",
+        ];
         for (const size of sizes) {
             const result = await imageClient.generate({
                 prompt: "A futuristic city skyline at sunset",

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

@@ -6,16 +6,24 @@ export interface DoubaoImageGenerationConfig {
     baseUrl?: string;
     headers?: Record<string, string>;
 }
+export type DoubaoImageSize = "2048x2048" | "2304x1728" | "1728x2304" | "2560x1440" | "1440x2560" | "2496x1664" | "1664x2496" | "3024x1296" | "1K" | "2K" | "4K";
 export interface DoubaoImageGenerationRequest {
     model?: string;
     prompt: string;
-    size?: string;
-    watermark?: boolean;
+    background?: "transparent" | "opaque" | "auto";
+    moderation?: "low" | "auto";
     n?: number;
-    quality?: "standard" | "hd";
+    output_compression?: number;
+    output_format?: "png" | "jpeg" | "webp";
+    partial_images?: number;
+    quality?: "standard" | "hd" | "low" | "medium" | "high" | "auto";
     response_format?: "url" | "b64_json";
+    size?: DoubaoImageSize;
+    stream?: false;
     style?: "vivid" | "natural";
+    image?: string | string[];
     user?: string;
+    watermark?: boolean;
 }
 export interface DoubaoImageData {
     url?: string;

package/dist/doubao-image-generation.js

@@ -27,25 +27,45 @@ class DoubaoImageGenerationClient {
         const requestBody = {
             model: request.model || "doubao-seedream-4-5-251128",
             prompt: request.prompt,
-            size: request.size || "2K",
+            size: request.size || "2048x2048",
             n: request.n || 1,
         };
-        // 添加可选参数
-        if (request.watermark !== undefined) {
-            requestBody.watermark = request.watermark;
+        // 添加可选参数（按照 OpenAI SDK 参数顺序）
+        if (request.background !== undefined) {
+            requestBody.background = request.background;
         }
-        if (request.quality) {
+        if (request.moderation !== undefined) {
+            requestBody.moderation = request.moderation;
+        }
+        if (request.output_compression !== undefined) {
+            requestBody.output_compression = request.output_compression;
+        }
+        if (request.output_format !== undefined) {
+            requestBody.output_format = request.output_format;
+        }
+        if (request.partial_images !== undefined) {
+            requestBody.partial_images = request.partial_images;
+        }
+        if (request.quality !== undefined) {
             requestBody.quality = request.quality;
         }
-        if (request.response_format) {
+        if (request.response_format !== undefined) {
             requestBody.response_format = request.response_format;
         }
-        if (request.style) {
+        if (request.stream !== undefined) {
+            requestBody.stream = request.stream;
+        }
+        if (request.style !== undefined) {
             requestBody.style = request.style;
         }
-        if (request.user) {
+        if (request.user !== undefined) {
             requestBody.user = request.user;
         }
+        // 添加图片输入参数（图文生图或多图融合）
+        if (request.image !== undefined) {
+            requestBody.image = request.image;
+        }
+        requestBody.watermark = request.watermark || false;
         const url = `${config_1.sdkConfig.getServerUrl()}/api/doubao-image-proxy/generate`;
         (0, log_1.logRequest)("POST", url, this.headers, requestBody);
         const response = await fetch(url, {

package/dist/index.d.ts

@@ -5,7 +5,7 @@
  * @see https://github.com/langchain-ai/langchainjs
  */
 import { BaseChatModel, BaseChatModelParams } from "./base";
-import { DoubaoImageGenerationClient, type DoubaoImageGenerationConfig, type DoubaoImageGenerationRequest, type DoubaoImageGenerationResponse } from "./doubao-image-generation";
+import { DoubaoImageGenerationClient, DoubaoImageSize, type DoubaoImageGenerationConfig, type DoubaoImageGenerationRequest, type DoubaoImageGenerationResponse } from "./doubao-image-generation";
 import { GeminiImageGenerationClient, type GeminiImageGenerationConfig, type GeminiImageGenerationRequest, type GeminiImageGenerationResponse } from "./gemini-image-generation";
 import { VideoGenerationClient, type VideoGenerationConfig, type VideoGenerationRequest, type ContentGenerationTaskID, type ContentGenerationTask } from "./video_generation";
 import { sdkConfig } from "./config";
@@ -19,9 +19,8 @@ export interface LangchainClientConfig {
     token?: string;
     headers?: Record<string, string>;
 }
-export { DoubaoImageGenerationClient, type DoubaoImageGenerationConfig, type DoubaoImageGenerationRequest, type DoubaoImageGenerationResponse, };
+export { DoubaoImageGenerationClient, type DoubaoImageGenerationConfig, type DoubaoImageGenerationRequest, type DoubaoImageGenerationResponse, type DoubaoImageSize, };
 export { GeminiImageGenerationClient, type GeminiImageGenerationConfig, type GeminiImageGenerationRequest, type GeminiImageGenerationResponse, };
-export { DoubaoImageGenerationClient as ImageGenerationClient, type DoubaoImageGenerationConfig as ImageGenerationConfig, type DoubaoImageGenerationRequest as ImageGenerationRequest, type DoubaoImageGenerationResponse as ImageGenerationResponse, };
 export { VideoGenerationClient, type VideoGenerationConfig, type VideoGenerationRequest, type ContentGenerationTaskID, type ContentGenerationTask, };
 export { sdkConfig };
 /**

package/dist/index.js

@@ -6,13 +6,12 @@
  * @see https://github.com/langchain-ai/langchainjs
  */
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.sdkConfig = exports.VideoGenerationClient = exports.ImageGenerationClient = exports.GeminiImageGenerationClient = exports.DoubaoImageGenerationClient = exports.ChatAnthropic = exports.ChatGoogleGenerativeAI = exports.ChatOpenAI = exports.BaseChatModel = exports.AIMessageChunk = exports.SystemMessage = exports.AIMessage = exports.HumanMessage = void 0;
+exports.sdkConfig = exports.VideoGenerationClient = exports.GeminiImageGenerationClient = exports.DoubaoImageGenerationClient = exports.ChatAnthropic = exports.ChatGoogleGenerativeAI = exports.ChatOpenAI = exports.BaseChatModel = exports.AIMessageChunk = exports.SystemMessage = exports.AIMessage = exports.HumanMessage = void 0;
 exports.createChatModel = createChatModel;
 const openai_1 = require("./chat_models/openai");
 const google_1 = require("./chat_models/google");
 const doubao_image_generation_1 = require("./doubao-image-generation");
 Object.defineProperty(exports, "DoubaoImageGenerationClient", { enumerable: true, get: function () { return doubao_image_generation_1.DoubaoImageGenerationClient; } });
-Object.defineProperty(exports, "ImageGenerationClient", { enumerable: true, get: function () { return doubao_image_generation_1.DoubaoImageGenerationClient; } });
 const gemini_image_generation_1 = require("./gemini-image-generation");
 Object.defineProperty(exports, "GeminiImageGenerationClient", { enumerable: true, get: function () { return gemini_image_generation_1.GeminiImageGenerationClient; } });
 const video_generation_1 = require("./video_generation");

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ai-world-sdk",
-  "version": "1.0.9",
+  "version": "1.0.10",
   "description": "TypeScript SDK for AI World Platform - Chat Models, Image Generation, and Video Generation",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",