manim-gpt / README.md
Alexey Mametyev
Merge pull request #6 from freQuensy23-coder/codex/update-setup-scripts-to-support-manimml
76ae8f7 unverified
---
title: Manim GPT - AI Video Creator
emoji: 🎬
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.32.0
app_file: demo.py
pinned: false
license: mit
---
# 🎬 Manim GPT - AI Video Creator
Create beautiful explanatory animations from simple text prompts using AI and Manim!
## Features
- πŸ€– **AI-Powered**: Uses Gemini AI to generate Manim code from natural language
- πŸŽ₯ **Automatic Rendering**: Creates high-quality MP4 videos
- 🎡 **Background Music**: Automatically adds background music to all videos
- πŸ”„ **Auto-Fix**: AI automatically fixes rendering errors and lets you iterate on the result
- ⚑ **Fast**: Medium quality rendering for quick results
## How to Use
1. Describe the concept you want to visualize (e.g. "Explain how neural networks work")
2. Click "Send" to start the generation process
3. The AI will create a scenario describing the animation
4. Press the **Next Step** button when you're ready to generate the Manim code
5. Wait for the video to render with background music
6. If you want changes, type a new message and the video will be regenerated
7. Download your animated explanation!
## Technical Details
- **AI Model**: Gemini 2.5 Flash Preview
- **Animation Engine**: Manim Community Edition
- **Video Processing**: MoviePy for audio mixing
- **Background Music**: Automatically loops/trims to match video duration
## Requirements
Set your Gemini API key as an environment variable:
```bash
export GEMINI_API_KEY="your_api_key_here"
```
Install dependencies (includes **manim-ml** for ML visualizations):
```bash
pip install -r requirements.txt
```
You can explore additional examples in [Google's ManimML code samples](https://github.com/google/manim-ml/tree/main/examples).