Stay Hungry,Stay Foolish!

Generative Manim

Generative Manim

https://generative-manim.vercel.app/

⚡️ How it works

Making video animations accessible to everyone.

Generative Manim is an open-source tool for generating animation videos from text. It is divided in three parts:

  • Generate Video: Generate the video from the text.
  • Render Engine: Render the video from the code.
  • Prompt Generator: Generate the code from the text.

The Generate Video part is the sum of the Prompt Generator and the Render Engine in a single step.

💡 Concept

Imagine a future where you can watch an animation video from a concept you have in your mind in a few seconds. That would be useful for teachers, students, and more. People that want to generate a video won't need to learn how to use a video editor, how to draw, or how to code. You just need to write a descriptive text.

That indeed is a future. And it is not that far. We can already start working on it.

Manim is a Python library for creating complex graphics and animations. The main advantage of Manim for GPT, it's that since it's a language for the LLM, it's more easy to generate proper code from a prompt.

💻 Models

The models we use to generate Manim code from text.

Generative Manim uses models to generate code from text. The first model we used was GPT-4 with a Zero Shot Learning strategy. Today we are using two models.

🤖 GPT-4o

GPT-4o is the next-generation model from OpenAI. We use a basic system prompt to generate the code. We add a few extra instructions to help the model understand the task.

📡 GPT-3.5 Fine-Tuned

We trained a GPT-3.5 model with a few examples of Manim code. The model is fine-tuned to generate Manim code from a prompt. This model is a little bit slower than GPT-4o.

💡 GPT-3.5 Fine-Tuned for Physics

A GPT-3.5 model fined-tuned to generate Manim for the Manim Physics plugin.

If you have an idea or want to help us, go to our Discord Community and let us know.

🚀 Contributors

Awesome people that are contributing to the project.

Generative Manim is an open source project. You can contribute to the project by suggesting new features, fixing bugs, or improving the documentation. Pull Requests are warmly welcome.

Remember you can also join our Discord Community to discuss new features, bugs or any other topic.

 

https://github.com/marcelo-earth/generative-manim

 

Generative Manim

 

🎨 GPT-4o powered generative videos. Concept. ⚡️ Join our Discord server here!


Preview

🚀 Concept

Generative Manim (GM) is a suite of tools that allows you to create videos with Manim using LLMs (Large Language Models) like GPT-4 or Claude. The idea is to enable anyone to create wonderful animations from text ✨.

It began as a prototype of a web app that uses GPT-4 to generate videos with Manim. The idea behind this project is taking advantage of the power of LLMs in programming, the understanding of human language and the animation capabilities of Manim to generate a tool that could be used by anyone to create videos. Regardless of their programming or video editing skills.

💻 Models

Models are the core of Generative Manim. A model is a way to convert text to code, that can later be rendered in a video.

NameDescriptionEnginePhase
GM GPT-4o Latest GPT model from OpenAI powered by a custom System Prompt GPT-4o
GM GPT-3.5 Fine Tuned First Fine-tuned model of GPT-3.5 GPT-3.5
GM GPT-3.5 Physics Fine Tuned Fine-tuned GPT-3.5 model trained to generate Physics animations GPT-3.5
GM Claude Sonnet Claude Sonnet 3 model from Sonnet adapted with our custom System Prompt claude-3-sonnet-20240229
GM Claude Sonnet 3.5 Claude Sonnet 3.5 model from Sonnet adapted with our custom System Prompt claude-3-5-sonnet-20240620

📡 New Models

If you want to suggest a new model, please open an issue in the repository or talk with us in our Discord server.

✨ Sponsors

Generative Manim is currently sponsored by The Astronomical Software Company.

🤲 Contributing

Generative Manim is an open source project.

If you want to be the author of a new feature, fix a bug or contribute with something new.

Fork the repository and make changes as you like. Pull requests are warmly welcome. Remember you can also join our Discord server to discuss new features, bugs or any other topic.

 

posted @   lightsong  阅读(44)  评论(0编辑  收藏  举报
相关博文:
阅读排行:
· 全网最简单!3分钟用满血DeepSeek R1开发一款AI智能客服,零代码轻松接入微信、公众号、小程
· .NET 10 首个预览版发布,跨平台开发与性能全面提升
· 《HelloGitHub》第 107 期
· 全程使用 AI 从 0 到 1 写了个小工具
· 从文本到图像:SSE 如何助力 AI 内容实时呈现?(Typescript篇)
历史上的今天:
2024-01-14 python client of influxdb v2
2024-01-14 sklearn.preprocessing + keras
2024-01-14 different python version + venv
2021-01-14 Working With Text Data of sklearn
千山鸟飞绝,万径人踪灭
点击右上角即可分享
微信分享提示