Transform Images to Editable 3D Models
PartPacker leverages NVIDIA's dual-volume packing AI to convert a single 2D image into inter-connected 3D parts ready for editing and 3D printing.

What is PartPacker
PartPacker represents a breakthrough in 3D generation technology, enabling novel workflows that were previously impossible with traditional single-mesh models.
- Single Image InputPartPacker requires only a single 2D RGB image to generate comprehensive 3D models, unlike competitors that need multiple viewpoints.
- Part-Based GenerationRevolutionary dual-volume packing strategy creates interconnected parts that can be individually manipulated.
- 3D Printing ReadyExport to STL/3MF formats optimized for multi-material 3D printing applications.
Why Choose PartPacker
Revolutionary 3D generation technology that creates editable, part-based models perfect for content creation and research.



How PartPacker Works
Generate editable 3D models from single images in three simple steps:
Key Features of PartPacker
Everything you need for advanced 3D model generation from single images.
Dual-Volume Packing
Revolutionary strategy that organizes parts into two complementary volumes for complete model generation.
Diffusion Transformer
Advanced DiT architecture ensures high-quality 3D model output with superior detail and accuracy.
Multi-Format Export
Export generated models to STL and 3MF formats optimized for 3D printing and editing applications.
Real-Time Processing
Efficient generation pipeline that processes single images into complete 3D models quickly.
Research-Grade Quality
Built on NVIDIA research with professional-grade output suitable for academic and commercial use.
Open Source
Available on GitHub and Hugging Face Hub with comprehensive documentation and model downloads.
PartPacker Technology Stats
Powered by cutting-edge AI research and technology.
Input Resolution
518×518
Pixels
VRAM Required
16GB+
Memory
Research Year
2024
Latest
PartPacker Applications
Discover how researchers and creators use PartPacker for innovative 3D generation projects.
Content Creation
Game Development
Generate editable 3D assets for games, films, and interactive media from concept art or reference images with individual part manipulation.
3D Printing
Manufacturing
Create printable models with separable components for multi-material printing and assembly applications in prototyping and production.
Academic Research
Computer Vision
Advance 3D generation research with PartPacker's innovative part-based approach and dual-volume packing methodology.
Product Design
Industrial Design
Transform 2D product concepts into editable 3D models for rapid prototyping and design iteration workflows.
Educational Tools
Learning Resources
Create interactive 3D models for educational content, allowing students to explore and manipulate object components.
Art & Animation
Creative Industries
Generate base 3D models from artwork for further refinement in animation and digital art production pipelines.
Frequently Asked Questions
Get answers to common questions about PartPacker's 3D generation technology.
What makes PartPacker different from other 3D generation tools?
PartPacker uses a revolutionary dual-volume packing strategy that generates interconnected, separable 3D parts from single images. Unlike traditional methods that create single-mesh models, PartPacker enables individual part manipulation for editing and animations.
What are the system requirements for running PartPacker?
PartPacker requires approximately 16GB+ VRAM for optimal performance, CUDA support, Python 3.8+, and PyTorch 2.0+. Professional-grade GPU hardware is recommended for best results.
What input image formats and resolutions does PartPacker support?
PartPacker accepts standard RGB image formats including JPG, PNG, and others. The recommended resolution is 518×518 pixels for optimal 3D generation quality.
Can I use PartPacker-generated models for commercial projects?
PartPacker is currently available for non-commercial use under NVIDIA's research license. For commercial usage and licensing options, please contact NVIDIA directly.
How can I access PartPacker for my projects?
You can access PartPacker through our interactive demo, download the source code from GitHub, or use the pre-trained models available on Hugging Face Hub.
What file formats can PartPacker export to?
PartPacker can export generated 3D models to STL and 3MF formats, which are optimized for 3D printing and support multi-material applications.
Ready to Transform Your Images?
Join researchers, creators, and developers worldwide who are pushing the boundaries of 3D generation technology.