Rodin3D Skills

A Skill for generating 3D models from image/multiview images or text prompts using the Hyper3D Rodin Gen-2 API.

Overview

This skill enables AI to generate high-quality 3D models through the Hyper3D Rodin Gen-2 API. It supports both image-to-3D and text-to-3D generation, with multiple output formats and quality tiers.

Features

• Generate 3D models from single or multiple images (up to 5 images)
• Generate 3D models from text descriptions
• Support for multiple output formats: GLB, USDZ, FBX, OBJ, STL
• Five generation tiers: Gen-2, Detail, Smooth, Regular, Sketch
• Automatic task status polling and result retrieval
• Advanced parameters for material, mesh mode, seed values, and more

Parameters

Basic Parameters

Advanced Parameters

Best Practices

Image Input

• Use high-resolution images (at least 512 x 512 pixels)
• Ensure good lighting and clear details
• Avoid cluttered backgrounds
• Use multiple angles for better results
• Supported formats: JPEG, PNG, WEBP
• File size limit: 16MB
• Resolution range: 512x512 to 4096x4096

Text Prompts

• Be specific and detailed in your descriptions
• Mention materials and textures
• Include lighting information
• Specify the style (realistic, cartoon, futuristic, etc.)
• Provide context for the object

Important Notes

Download links expire in 10 minutes: API result URLs are temporary. Download 3D models immediately after generation completes. Do not store or cache the URLs themselves.

API Endpoints

Example Prompts

Here are some example text prompts for 3D generation:

1. "A detailed 3D model of a medieval castle"
2. "A futuristic cityscape with flying cars"
3. "A realistic 3D model of a cat sitting on a couch"
4. "An abstract sculpture made of glass and metal"
5. "A 3D model of a cozy cottage in the woods"

Documentation

For detailed documentation, see:

• SKILL.md - Complete skill documentation
• assets/examples/README.md - Example usage

Links

• Hyper3D Website
• API Dashboard
• Hyper3D API Documentation

Support

For issues or questions:

• Submit a GitHub Issue
• Refer to SKILL.md for detailed documentation
• Visit Hyper3D official documentation

Author: HyperHuman
Version: 1.0.0
Tags: 3D Asset, Hyper3D, Rodin Gen-2, API, 3D Model, Image-to-3D, Text-to-3D

Installation

Prerequisites

• Claude Code CLI installed
• Hyper3D API Key (obtain from https://hyper3d.ai/api-dashboard)
• Python 3.7+ (for backend scripts)

Method 1: Install from Marketplace (Recommended)

The easiest way to install this skill is through the Claude Code plugin marketplace:

1. Add the marketplace (one-time setup):

   Copy

   /plugin marketplace add DeemosTech/rodin3d-skills

2. Install the plugin:

   Copy

   /plugin install rodin3d-skill@rodin3d-skills

3. Install Python dependencies:

   Copy

   pip install requests pillow

4. Set your API key:

   Copy

   export HYPER3D_API_KEY=your_api_key_here

   Or create a .env file in your project:

   Copy

   echo 'HYPER3D_API_KEY=your_api_key_here' > .env
   echo '.env' >> .gitignore

That's it! The skill is now available as /rodin3d-skill in Claude Code.

Method 2: Manual Installation

If you prefer to install manually or want to develop locally:

1. Clone this repository:

   Copy

   git clone https://github.com/DeemosTech/rodin3d-skills.git
   cd rodin3d-skills

2. Install Python dependencies:

   Copy

   cd skills/rodin3d-skill
   pip install -r requirements.txt

3. Add the local marketplace:

   Copy

   /plugin marketplace add /path/to/rodin3d-skills
   /plugin install rodin3d-skill@rodin3d-skills

4. Set your API key (same as Method 1, step 4)

Verify Installation

After installation, verify the skill is available:

/plugin list

You should see rodin3d-skill in the list of enabled plugins.

Updating the Plugin

To update to the latest version:

/plugin marketplace update rodin3d-skills
/plugin update rodin3d-skill@rodin3d-skills

Usage

Using the Skill

Once installed, you can interact with this skill through:

Generate from image:

Please generate a 3D model from this image: [provide image path]

Generate from text:

Please generate a 3D model of a medieval castle with high quality

Generate from multiple images:

Please generate a 3D model from these images: [provide multiple images]

Direct Script Usage

You can also use the Python scripts directly:

python scripts/generate_3d_model.py \
  --image path/to/image.jpg \
  --geometry-file-format glb \
  --quality medium \
  --output ./output

python scripts/generate_3d_model.py \
  --prompt "A detailed 3D model of a medieval castle" \
  --geometry-file-format glb \
  --quality high \
  --output ./output

Configuration

Generation Tiers

Output Formats

• GLB: Recommended for web and most 3D software
• USDZ: For Apple platforms
• FBX: For game engines and 3D modeling software
• OBJ: Universal 3D format
• STL: For 3D printing

Quality Levels

• High: Best quality, longer processing time
• Medium: Balanced quality and speed
• Low: Faster processing, lower quality
• Extra-low: Fastest processing, minimal quality

Troubleshooting

API Key Issues

If you encounter authentication errors:

1. Verify your API key is set correctly
2. Check that the key is valid and active
3. Ensure you have sufficient API credits

Generation Failures

If generation fails:

1. Check image format and size requirements
2. Verify prompt is not empty or too short
3. Ensure API service is available
4. Review error messages for specific issues

License

MIT License - See LICENSE file for details