A professional-grade prompt engineering workbench for Google's Gemini models. Built with React, TypeScript, and Vite, this application provides a powerful interface for interacting with Gemini 2.5 and 3.0 models with advanced control over generation parameters.
- Multi-Model Support: Access
Gemini 2.5 Flash,Gemini 3.0 Pro Preview, andGemini 2.5 Flash (Thinking). - Multimodal Chat: Seamlessly send text and images (multiple attachments supported).
- Streaming Responses: Real-time typewriter-style response generation.
- Contextual History: Maintains conversation context for natural interactions.
- System Instructions: Define custom personas and behavior rules.
- Parameter Tuning:
- Temperature: Control creativity (0.0 - 2.0).
- Top-K & Top-P: Fine-tune vocabulary sampling.
- Thinking Budget: Configure token budgets (1024 - 8192) for reasoning models.
- Search Grounding: Toggle Google Search integration for real-time data.
- Professional Dark Mode: Developer-focused interface with syntax highlighting.
- Markdown Rendering: Rich text formatting for code, tables, and lists.
- Responsive Design: Fully functional on desktop and mobile devices.
- Frontend: React 19, TypeScript, Vite
- Styling: TailwindCSS, Lucide React
- AI Integration: Google GenAI SDK
- Testing: Vitest, React Testing Library
- Deployment: Netlify, Docker support
- Node.js (v20 or higher)
- A Google Gemini API Key
-
Clone the repository
git clone https://github.com/madhan-200/Chota-Architecture.git cd Chota-Architecture -
Install dependencies
npm install
-
Configure Environment Create a
.env.localfile in the root directory:GEMINI_API_KEY=your_api_key_here
-
Run Locally
npm run dev
Access the app at
http://localhost:3000.
Run the unit test suite to verify components:
npm testThe project is configured for seamless deployment on Netlify.
npm run build
npx netlify deploy --prodBuild and run the containerized application:
docker build -t chotta-architect .
docker run -p 80:80 chotta-architectMIT License