VisiGen

Turn words into vivid images — right on your desktop.

By making this change, your Hugging Face badge should now display correctly on your project's README.

View Demo · Report Bug · Request Feature

---

✨ Overview

VisiGen is a JavaFX desktop application that converts natural-language prompts into AI-generated images using a Stable Diffusion–style model hosted on Hugging Face. It’s designed for devs, designers, and AI enthusiasts who want a lightweight, offline-friendly front-end without the browser overhead.

• ⚡ Fast – one-click prompt ➜ preview in seconds
• 🎨 Customizable – tweak prompt templates, aspect ratios, inference steps
• 🖥️ Desktop-native – smooth JavaFX UI, no Electron bloat
• 🔌 Open API – swap in any Hugging Face text-to-image endpoint with a single env var

Status: MVP complete · Feature roadmap in progress — PRs welcome!

---

🚀 Quick Start

# 1) Clone the repo
git clone https://github.com/dave21-py/VisiGen.git
cd VisiGen

# 2) Create and activate a virtual env (optional but recommended)
python3 -m venv venv            # Windows:  py -m venv venv
source venv/bin/activate        # Windows:  venv\Scripts\activate

# 3) Install helper scripts (gradle wrapper, env loader) 
pip install -r tools/requirements.txt

# 4) Set your API key
cp .env.example .env
# edit .env and add:
#   HF_API_KEY=hf_xxxxxxxxxxxxxxxxxxxxx

# 5) Build & run (macOS/Linux)
./gradlew clean run
# Windows PowerShell
.\gradlew.bat clean run
****

```bash
git clone https://github.com/dave21-py/huggingface-image-generator.git

Gradle Setup

gradle init --type java-application --dsl groovy --package app --test-framework junit-jupiter --use-defaults --overwrite 

Gradle commands

gradle clean build

gradle run

JavaFX Setup

Add a line to app/build.gradle

plugins {
    id 'application'
    id 'jacoco'
    id 'org.openjfx.javafxplugin' version '0.1.0'
}

Add a new block to the bottom of app/build.gradle:

javafx {
    version = "23.0.1"
    modules = [ 'javafx.controls', 'javafx.fxml', 'javafx.media' ]
}

👀 Demo Prompts

Prompt 1:

An ethereal, three-dimensional neural network visualizing real-time sentiment analysis from global data streams. Nodes pulse with color-coded light blue for positive, red for negative, yellow for neutral, interconnected by shimmering, flowing data pathways. Abstract, intricate, glowing, high detail, cinematic lighting, bokeh background.

Prompt 2:

Interior of a futuristic autonomous vehicle. The windshield is a fully transparent, adaptive display showing AI-curated points of interest, navigation overlays, and personalized entertainment. Passengers interact with the system using natural language and subtle hand gestures. Luxurious, ambient lighting, wide-angle shot, detailed interior.

Prompt 3:

A sunset over the ocean with glowing waves

System Requirements

Both MacOS and Windows supported.

🚀 Citation

If you find our work useful, please consider citing:

title={VisiGen},
  author={David Geddam},
  year={2025}
}

📭 Contact

For any questions, please raise a concern or contact us at

• David Geddam linkedin.com