Skip to content

Spell by Spline

Generate a full 3D scene from a single image

Visit official site

What Spell is

The model is described as an AI model from Spline that turns a single image into a complete 3D scene. The material states diffusion models trained on real and synthetic data produce multi-view consistent worlds in minutes.

What you can do with it

  • Upload one image as input
  • Choose an output such as video, image sequence or volume
  • Adjust the camera path and other settings

Rendering notes

  • Gaussian splatting and neural radiance fields are supported
  • Reflection, refraction and surface roughness can be simulated
  • Volume representations can be converted to meshes

Pros & cons

✓ What we like

  • Internal volume representations can become meshes
  • People, objects and environments are supported

! What to watch out for

  • The material describes the model as still in development
  • Scene generation follows one uploaded image

FAQ

What is the input?

The material states a single uploaded image.

Which rendering methods are named?

Gaussian splatting and neural radiance fields.

Last reviewed: 2026-11-15

More AI design tools tools

View all →

How we review