Out of sheer curiosity to test the limits of the new Google Gemini models, I built Amaris—a full-stack SaaS application that lets you generate images with specific artistic styles using a simple chat interface. No complex workflows, just upload a reference image, type a prompt, and let the AI do the heavy lifting!
It's built with the "Better-T-Stack" (Bun, Hono, React 19) and uses Google's Gemini 2.5 Flash Image model via the Vercel AI Gateway. If you're interested in building modern AI apps, learning about monorepos, or just want to see how Hono handles AI streams, this project is a great sandbox.
Demo video:
https://x.com/devguy_007/status/1997485877325394061?s=20
Check it out here:
🔗 [https://github.com/oyeolamilekan/amaris-app]
✨ Features
- 💬 Chat-Based Interface – Generate images in a conversational flow.
- 🎨 Style Transfer – Upload a reference image (like a sketch or painting) and the AI mimics that style.
- ⚡ Blazing Fast Backend – Powered by Hono and running on Bun.
- 🧠 Smart Vision – Uses Gemini to "see" your style references before generating.
- 🔐 Full Auth System – Secure login with Better-Auth.
- 💳 Credit System – Built-in logic for managing user credits and usage limits.
- 🌑 Modern UI – Dark mode, Tailwind CSS 4, and shadcn/ui components.
🧰 The Complete Tech Stack
I threw the kitchen sink of modern web dev at this project. Here is everything powering Amaris:
Core & Runtime
| Tech | Description |
|---|---|
| Bun | Ultra-fast JavaScript runtime & package manager |
| TypeScript | For type safety across the entire monorepo |
Frontend (apps/web)
| Tech | Description |
|---|---|
| React 19 | The latest version of the library |
| React Router 7 | For declarative, file-based routing |
| Tailwind CSS 4 | The newest utility-first CSS engine |
| shadcn/ui | Beautiful, accessible UI components |
| TanStack Query | For async state management and caching |
| TanStack Form | For handling complex forms with ease |
| Lucide React | Clean, consistent icons |
| Sonner | For those nice toast notifications |
Backend (apps/server)
| Tech | Description |
|---|---|
| Hono | Lightweight, standards-based web framework |
| Zod | Schema validation for API requests |
| Hono/Zod-Validator | Middleware to validate incoming data |
Data & Authentication
| Tech | Description |
|---|---|
| PostgreSQL | The relational database |
| Drizzle ORM | TypeScript-first ORM for type-safe SQL |
| Better-Auth | Comprehensive authentication (Email/Pass, Sessions) |
AI & Services
| Tech | Description |
|---|---|
| Google Gemini | 2.5 Flash Image (Generation) & 2.5 Flash (Vision) |
| Vercel AI SDK | AI Gateway for caching, rate-limiting, and analytics |
| Cloudinary | For storing and optimizing generated images |
| Polar.sh | For handling subscriptions and payments |
🚀 Getting Started Locally
Here's how to get Amaris running on your machine:
1. Clone the Repo
git clone https://github.com/oyeolamilekan/amaris-app
cd amaris
2. Install Dependencies
We use Bun for speed!
bun install
3. Set Up Environment
You'll need a PostgreSQL DB and a Google AI Studio key. Create your .env files:
apps/server/.env
DATABASE_URL=postgresql://user:pass@localhost:5432/amaris
GOOGLE_GENERATIVE_AI_API_KEY=your_gemini_key
CORS_ORIGIN=http://localhost:5173
CLOUDINARY_CLOUD_NAME=your_cloud_name
POLAR_ACCESS_TOKEN=your_polar_token
4. Push the Database Schema
bun run db:push
5. Start the Dev Server
bun run dev
Visit http://localhost:5173 to start generating!
🧪 How It Works
- Upload a Style: You drop an image into the chat (e.g., a pixel art character).
- Vision Analysis: The backend uses Gemini's vision capabilities to understand the "vibe" of that image.
- Prompting: You type "A futuristic city."
- Generation: The app combines your prompt with the style reference and sends it through the Vercel AI Gateway to Google Gemini.
- Result: Seconds later, you get a futuristic city rendered in pixel art style!
💭 Why I Built This
I wanted to move beyond simple "text-to-image" scripts and build a real, production-ready architecture. I was specifically interested in how Hono (the backend framework) pairs with React 19 in a monorepo setup, and how to integrate payment flows like Polar.sh alongside complex AI logic.
Plus, the new Gemini 2.5 models are incredibly fast and cheap, making them perfect for experimenting with style transfer without breaking the bank.
🧑💻 Try It, Fork It, Break It!
This project is open-source and meant to be hacked on. Feel free to clone the repo, swap out the database, try a different AI model, or turn it into a comic book generator.
I'd love to see what you build with it!
GitHub: [https://github.com/oyeolamilekan/amaris-app]
🙌 Let's Connect
If you try it out or have questions about the stack, drop a comment or tag me!
Happy hacking! 👨🏽💻✨
Top comments (0)