Stable Zero123
Stable Zero123 generates views conditioned on an image and camera position. The release includes original and commercially licensed variants with different training-data restrictions.
Model design and access
The original checkpoint includes objects with non-commercial licenses. Zero123C was trained only with CC-BY and CC0 objects. The card traces the image model to Lambda Labs' image-variation diffusion model. Text-to-3D requires an image-generation stage before reconstruction.Stable Zero123 official model card
Practical use
Practical guidance. Choose the checkpoint before building a workflow around it, because the two variants do not have the same usage terms. The model supplies proposed views; mesh optimization is a separate operation. Check whether a newer direct mesh model better matches the required output before accepting an iterative reconstruction process.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | Stable Zero123Cstable_zero123stable_zero123_c |
| Method | View-conditioned latent diffusion; mesh creation uses an additional SDS workflow.Stable Zero123 official model card |
| Original license | Non-Commercial Research Community License.Stable Zero123 official model card |
| Zero123C license | Stability AI Community License.Stable Zero123 official model card |
Related models
- Stable Fast 3D
Stability AI · Stable 3D
Stable Fast 3D reconstructs a textured mesh from a single object image.
Launch date not confirmedAPI · Downloadable - Stable Video 3D
Stability AI · Stable Video
Stable Video 3D adapts Stable Video Diffusion to generate orbital views of an object from one image.
Launch date not confirmedDownloadable