NanoLayer Studio is a professional-grade, web-based image editor powered by Google's Gemini generative AI models. It combines traditional layer-based editing workflows with cutting-edge AI capabilities for generation, modification, and analysis.
- Full Layer System: Manage multiple layers with visibility toggles, opacity controls, and z-index ordering.
- PSD Support: Import and Export Adobe Photoshop (.psd) files, preserving layer structure.
- Standard Image Support: Import/Export PNG, JPG, and WebP formats.
- Dual Model Support:
- Gemini 2.5 Flash Image (Nano Banana): Fast, efficient model for quick edits.
- Gemini 3 Pro Image Preview: High-fidelity model for complex generation and reasoning.
- Generative Edit: Use text prompts to modify existing images (e.g., "Add sunglasses", "Change background to space").
- Region Selection (Masking): Select specific areas of the canvas to apply AI edits locally without affecting the rest of the image.
- Style Reference: Select a specific layer to serve as a "Style Reference" for your generations (Multimodal editing).
- System Instructions: Define global style guides (e.g., "Cyberpunk style", "Oil painting") to ensure consistent outputs.
- Deep Understanding: Utilize Gemini 3 Pro to analyze layers and get detailed descriptions, style breakdowns, or answers to questions about the image content.
- Internationalization: Full support for English and Chinese (中文) interfaces.
- Custom API Key: Use the built-in system key or provide your own Google GenAI API Key via Settings (stored locally).
- Responsive Workspace: Infinite canvas feel with zoom and pan adaptation (auto-fit).
- Frontend: React 18, TypeScript, Vite
- Styling: Tailwind CSS
- AI SDK: Google GenAI SDK (
@google/genai) - Image Processing: HTML5 Canvas API,
ag-psdfor Photoshop file handling
-
Clone the repository
git clone https://github.com/dyue708/nanolayer cd nanolayer -
Install Dependencies
npm install
-
Run Development Server
npm run dev
-
Build for Production
npm run build
To use the AI features, you need a valid Google GenAI API Key.
- You can create a
.envfile in the root directory:API_KEY=your_api_key_here - Or enter it directly in the application Settings menu (gear icon).
NanoLayer Studio 是一款专业级的网页图像编辑器,由 Google Gemini 生成式 AI 模型驱动。它结合了传统的图层编辑工作流与尖端的 AI 生成、修改和分析能力。
- 全功能图层系统:支持多图层管理,具备可见性切换、不透明度控制和 Z 轴排序功能。
- PSD 支持:支持导入和导出 Adobe Photoshop (.psd) 文件,并保留图层结构。
- 标准图片支持:支持导入/导出 PNG, JPG, 和 WebP 格式。
- 双模型支持:
- Gemini 2.5 Flash Image (Nano Banana):快速、高效,适合快速编辑。
- Gemini 3 Pro Image Preview:高保真模型,适合复杂的生成和逻辑推理任务。
- 生成式编辑:使用文本提示词修改现有图片(例如:“添加墨镜”,“把背景换成太空”)。
- 区域选择 (蒙版):选择画布的特定区域,仅对该区域应用 AI 编辑,不影响图片的其他部分。
- 风格参考:选择一个特定图层作为“风格参考”,实现多模态编辑与风格迁移。
- 系统提示词:定义全局风格指南(例如:“赛博朋克风格”,“油画风格”)以确保输出风格的一致性。
- 深度理解:利用 Gemini 3 Pro 分析图层,获取详细的描述、风格拆解,或回答关于图片内容的具体问题。
- 国际化:完全支持英文和中文(Chinese)界面切换。
- 自定义 API Key:支持使用内置系统 Key,或在“设置”菜单中输入您自己的 Google GenAI API Key(仅存储在本地)。
- 响应式工作区:具备无限画布体验,支持自动适应缩放和平移。
- 前端: React 18, TypeScript, Vite
- 样式: Tailwind CSS
- AI SDK: Google GenAI SDK (
@google/genai) - 图像处理: HTML5 Canvas API,
ag-psd(用于处理 Photoshop 文件)
-
克隆仓库
git clone https://github.com/dyue708/nanolayer cd nanolayer -
安装依赖
npm install
-
运行开发服务器
npm run dev
-
构建生产版本
npm run build
要使用 AI 功能,您需要一个有效的 Google GenAI API Key。
- 您可以在根目录创建一个
.env文件:API_KEY=your_api_key_here - 或者直接在应用程序的 设置 菜单(齿轮图标)中输入。
Powered by Google Gemini API