# DearPyGui — High-Performance Python GUI Framework with GPU Rendering

> A fast GPU-accelerated Python GUI toolkit built on Dear ImGui for data science, debugging, and interactive developer tools.

## Install

Save as a script file and run:

# DearPyGui — High-Performance Python GUI Framework

## Quick Use
```bash
pip install dearpygui
python -c "import dearpygui.dearpygui as dpg; dpg.create_context(); dpg.create_viewport(title='Demo'); dpg.setup_dearpygui(); dpg.show_viewport(); dpg.start_dearpygui(); dpg.destroy_context()"
```

## Introduction
DearPyGui brings Dear ImGui's immediate-mode rendering model to Python with native GPU acceleration. It targets developers who need responsive, interactive UIs for data visualization, tooling, and prototyping without the overhead of traditional widget frameworks.

## What DearPyGui Does
- Renders UI elements via GPU (DirectX, OpenGL, Metal) for smooth performance
- Provides 70+ widgets including plots, tables, drawlists, and node editors
- Supports real-time data plotting with thousands of data points at 60 FPS
- Offers a callback-driven API with simple Python function bindings
- Runs cross-platform on Windows, macOS, and Linux

## Architecture Overview
DearPyGui wraps a custom C++ rendering core that extends Dear ImGui with retained-mode convenience atop the immediate-mode engine. The Python layer communicates through CPython bindings so all heavy lifting stays in compiled code. A viewport manager handles OS window creation, input routing, and the render loop.

## Self-Hosting & Configuration
- Install from PyPI: `pip install dearpygui`
- Requires Python 3.8+ and a GPU-capable display
- Configure viewport size, position, and VSync at creation time
- Themes and fonts can be customized via the theming API
- No external config files needed; everything is set programmatically

## Key Features
- GPU-accelerated rendering for data-heavy interfaces
- Built-in node editor for visual programming UIs
- Live-updating plots that handle high-frequency data streams
- Full DPI-aware scaling and multi-viewport support
- Extensive callback system with item-level event handling

## Comparison with Similar Tools
- **Tkinter** — ships with Python but lacks GPU rendering and modern aesthetics
- **PyQt / PySide** — feature-rich but carries complex licensing and large dependencies
- **Kivy** — targets touch and mobile UIs rather than desktop tooling
- **Streamlit** — web-based, not suitable for latency-sensitive desktop UIs
- **egui (Rust)** — similar immediate-mode model but requires Rust

## FAQ
**Q:** Does DearPyGui support web or mobile deployment?
A: No. It is desktop-only (Windows, macOS, Linux) and requires a local GPU context.

**Q:** Can I use DearPyGui for production applications?
A: Yes, but it is designed primarily for tooling, dashboards, and developer UIs rather than consumer-facing apps.

**Q:** How does it compare to Dear ImGui in C++?
A: DearPyGui uses Dear ImGui internally but adds a retained-mode Python API, built-in plotting, and node editor support.

**Q:** Is there a drag-and-drop GUI builder?
A: There is no official visual builder. UIs are defined programmatically in Python.

## Sources
- https://github.com/hoffstadt/DearPyGui
- https://dearpygui.readthedocs.io

---
Source: https://tokrepo.com/en/workflows/asset-ea45d519
Author: Script Depot