Zero-1-to-3 (Zero123)

Foundational open model for seeing an object from new angles given one image

AI 3D Model GeneratorsLight codeFreeFree tier
DeveloperColumbia University; Toyota Research Institute
CategoryAI 3D Model Generators
Typeopen model
PlatformsLinux; Windows
Coding requiredLight code
Pricing modelFree
Free tierYes

Zero-1-to-3, also called Zero123, is Columbia University's open source diffusion model that generates novel views of an object from a single image and a chosen camera viewpoint. Its generated views feed 3D reconstruction pipelines, its checkpoints are public, and the official code is released under the MIT license.

Key features

AI features

Visit the Zero-1-to-3 (Zero123) website

More AI 3D Model Generators

3D AI Studio

Multi model studio for generating 3D assets with many AI engines in one place

AI 3D Model GeneratorsNo codingFree tier

Basic $19/month; Studio $29/month; Pro $149/month; one time credit packs from $34

3DPresso

Video to 3D extraction that rebuilds real objects as textured models

AI 3D Model GeneratorsNo codingFree tier

Pricing: Freemium

Adobe Project Neo (Beta)

Live Adobe beta for building dimensional designs from 3D shapes and text

AI 3D Model GeneratorsNo codingFree tier

Pricing: Free

Agisoft Metashape

Desktop photogrammetry software that turns photos into point clouds and textured meshes

AI 3D Model GeneratorsNo coding

Standard Edition $179; Professional Edition $3,499