Skip to content
Pricing: Free
Verified: Yes
Rating: 4.5/5

NVIDIA research model that generates textured, production-ready 3D assets with PBR materials from text or image inputs in around two minutes.

Category

3D Model

View all 3D Model tools
Verified Selection
Updated Recently
Community Reviewed

Pricing

Edify 3D is a research model published by NVIDIA Research and accessible through its project page at no cost. Commercial implementations of the Edify architecture are available through Shutterstock's Generative 3D service and Getty Images' Generative AI platform, both of which operate under their own subscription and licensing terms. The NVIDIA NIM microservice preview for Edify was discontinued in June 2025.

PlanDetails
FreeThe research model weights, project documentation, and web demo are publicly accessible at no cost through the NVIDIA Research project page and associated open repositories.
PaidCommercial access to Edify 3D architecture is available through Shutterstock's Generative 3D service and Getty Images' Generative AI platform, each operating under their own subscription and licensing terms. Enterprise API access via NVIDIA Cloud is subject to Tencent Cloud pricing where applicable.

What is Edify 3D?

Quick Summary

Edify 3D is a research AI model developed by NVIDIA that generates high-quality 3D assets from text prompts or reference images using a multi-view diffusion and mesh reconstruction pipeline. It is designed for 3D artists, game developers, VR content creators, and designers who need to produce detailed, production-ready 3D objects without manual modeling. The model outputs structured quad meshes with 4K textures, PBR materials, and UV maps in approximately two minutes per asset.

Edify 3D is a foundation model published by NVIDIA Research that converts text descriptions or reference images into fully textured, structured 3D assets. The pipeline operates in two main stages: first, a set of multi-view diffusion models generates RGB images and surface normal maps of the target object from multiple camera angles simultaneously; second, a Transformer-based reconstruction model uses those multi-view observations to predict geometry, texture, albedo, and PBR material channels, then extracts a clean quad mesh via isosurface processing. The result is a 3D asset with well-organized mesh topology, automatically generated UV maps, and up to 4K physically based rendering textures, completed in approximately two minutes from input to output. Assets use the OpenUSD format, enabling direct import into NVIDIA Omniverse USD Composer and compatible 3D pipelines. Edify 3D has been adopted in production workflows through partnerships with Shutterstock and Getty Images, both of which built commercially licensed generative 3D services on the Edify architecture. Game developers use it to rapidly prototype environment props and secondary scene objects. Architects and interior designers apply it to generate furniture and material variants for visualization. VR and XR studios use the model to quickly populate immersive scenes with custom assets. The technology was demonstrated publicly at SIGGRAPH 2024, where NVIDIA researchers used it live to generate dozens of distinct 3D objects—including rocks, cacti, and decorative items—within a few minutes as part of a full scene composition workflow. Edify 3D produces quad meshes with clean, editable topology that integrates well into standard CG and game engine pipelines, and its 4K PBR output reduces the need for post-generation texture work. However, as a research model, the hosted NIM microservice preview was discontinued in June 2025, meaning direct access now depends on the research project page and any derivative commercial services built on the architecture. Generated asset complexity is suited to individual objects and props rather than full environments, and users requiring scene-level generation may need to combine Edify 3D with complementary tools such as Edify 360 HDRi for backgrounds and USD Layout for scene assembly. Explore this option.

Associated Tags

text to 3D, image to 3D, 3D asset generation, PBR materials, multi-view diffusion, NVIDIA AI, OpenUSD, game asset generation

Key Features

Text-to-3D asset generation
Image-to-3D mesh reconstruction
Automatic 4K PBR texture output
Multi-view surface normal prediction
Structured quad mesh topology
OpenUSD format compatibility
NVIDIA Omniverse pipeline integration

Real Use Cases

How professionals leverage Edify 3D – NVIDIA AI 3D Asset Generation Model

Edify 3D – NVIDIA AI 3D Asset Generation Model use cases
  • A game developer inputs a text description of a decorative prop and receives a textured, game-ready quad mesh in about two minutes for direct import into Unreal Engine.
  • A VR studio uses reference images of real-world objects to generate 3D asset variants for populating an immersive virtual environment without manual modeling.
  • An architect generates multiple furniture and fixture variants from text prompts to quickly explore design options in an interior visualization scene inside NVIDIA Omniverse.
  • A product designer uses Edify 3D to create a draft 3D model from a 2D concept sketch, accelerating early-stage prototyping before committing to manual refinement.
  • A creative team at a media company uses an Edify-powered API through Shutterstock's Generative 3D service to produce custom secondary scene assets for an advertising campaign.
  • A researcher in computer vision uses Edify 3D's multi-view output pipeline as a reference architecture to study multi-view diffusion approaches for 3D reconstruction.

Editor's Verdict

Official Review
Edify 3D is a technically rigorous NVIDIA research model that meaningfully reduces the time and skill required to produce textured, production-ready 3D assets from text or image inputs, with clean quad mesh output and PBR material support. Its main limitation is that the direct NVIDIA-hosted service is no longer available, leaving users reliant on the research project page or derivative commercial platforms for access.
4.5 / 5.0
Editor Rating

Reviewed by Sohail Akhtar

Lead Editor & Founder

Pros

What we like

  • Produces structured quad meshes with 4K PBR textures and clean UV maps in approximately two minutes, making outputs directly usable in professional CG and game engine pipelines.
  • OpenUSD format support enables seamless integration with NVIDIA Omniverse and other USD-compatible 3D tools without additional conversion steps.
  • The architecture underpins commercially licensed services from Shutterstock and Getty Images, demonstrating real-world production viability beyond a research prototype.

Cons

Limitations

  • The NVIDIA-hosted NIM microservice preview was discontinued in June 2025, so direct access to the model now depends on the research page or third-party commercial services built on the architecture.
  • Edify 3D is designed for individual object and prop generation rather than full scene or environment creation, requiring additional tools for complete scene assembly workflows.

Target Audience

Who should use Edify 3D?

game developers3D artists and modelersVR and XR content creatorsarchitects and interior designersproduct and industrial designersresearchers in generative 3D AI
Freemium
Hunyuan World 1.0

Hunyuan World 1.0

Open-source AI model by Tencent that generates explorable, interactive 3D worlds from text or image inputs using panoramic scene reconstruction.

Freemium
CgDream

CgDream

Browser-based AI platform that generates images and 3D models from text or photos using Flux and Rodin models, with 300+ style filters and multiple export formats.

Free
Marble by World Labs

Marble by World Labs

Multimodal AI world model by World Labs that generates persistent, navigable 3D environments from text, images, video, or 3D layouts, with in-scene editing and Gaussian splat, mesh, and video export.

Free
Meshy AI

Meshy AI

Browser-based AI 3D model generator with text-to-3D, image-to-3D, AI texturing, auto-rigging, and animation tools, used by 5M+ creators and recognized by a16z as the top AI tool among game developers.

Frequently Asked Questions

What is Edify 3D?
Edify 3D is a foundation AI model developed by NVIDIA Research that generates structured, textured 3D assets from text descriptions or reference images using a multi-view diffusion and mesh reconstruction pipeline.
How does Edify 3D work?
The model generates RGB and surface normal images of the target object from multiple camera angles using diffusion models, then uses a Transformer-based reconstruction model to produce a quad mesh with 4K PBR textures and UV maps.
Is Edify 3D free to use?
The research model and project page are publicly accessible at no cost. However, the NVIDIA NIM microservice preview was discontinued in June 2025; commercial access is available through Shutterstock and Getty Images services built on the Edify architecture.
What file formats does Edify 3D output?
Edify 3D outputs assets in the OpenUSD format with structured quad meshes, making them directly compatible with NVIDIA Omniverse USD Composer and other USD-based 3D pipelines.
Who should use Edify 3D?
Edify 3D is best suited for game developers, 3D artists, VR content creators, architects, and product designers who need to quickly generate textured, production-ready 3D objects from text or image inputs.
How long does Edify 3D take to generate a 3D asset?
Edify 3D generates a complete 3D asset, including geometry, UV maps, and 4K PBR textures, in approximately two minutes from a text or image input.