README.md

metadata

title: Anim Lab AI
emoji: 🔥
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.50.0
app_file: app.py
pinned: true
tags:
  - building-mcp-track-creative
  - building-mcp-track-consumer
  - mcp
  - gradio
  - claude
  - education
  - mathematics
  - algorithm
  - ai
  - visualization
  - education
short_description: Turn boring math into engaging animation videos

🎬 Anim Lab AI (MCP)

Equip AI Agents with the power to visualize complex logic using Manim.

📖 Overview

This is an MCP Server built with Gradio that allows AI agents to generate, execute, and render Manim (Mathematical Animation Engine) videos on the fly.

While LLMs are great at text, they struggle to visualize dynamic concepts. This tool bridges that gap by allowing an agent to write Python/Manim code and immediately receive a rendered video file.

Key Features

• Text-to-Video: Describe a math concept (e.g., "Explain Bubble Sort"), and the AI writes the code and renders the video.
• AI Code Editor: Refine animations using natural language instructions (e.g., "Move the circle to the left").
• MCP Compliant: Fully functional as a backend tool for Claude Desktop or other MCP clients.
• Powered by Anthropic Claude: Uses anthropic/claude-4.5-sonnet for high-precision code generation.

🚀 Quick Start

1. Using the Web UI

1. Type a prompt in the "Describe your animation" box.
2. Click Generate & Render.
3. Watch your code turn into a video!

2. Connect via MCP (Streamable HTTP)

To add this MCP to clients that support Streamable HTTP, add the following configuration to your MCP config:

{
  "mcpServers": {
    "gradio": {
      "url": "https://mcp-1st-birthday-anim-lab-ai.hf.space/gradio_api/mcp/"
    }
  }
}

⚙️ Tech Stack

• Frontend/Server: Gradio (MCP Mode)
• Animation Engine: Manim Community (v0.19)
• LLM Inference: Fal AI (Claude 4.5 Sonnet)

Live Demo & Demo Video

• Live Demo: Anim Lab AI - Huggingface
• Demo Video: Youtube

Social Media Post

• Huggingface Post: Introducing Anim Lab AI - HF
• Twitter/X: Introducing Anim Lab AI - X

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference