# Structured 3D Latents for Scalable and Versatile 3D Generation

*Source* · `src-xiang-2024-trellis` · status: **n/a**

| field | value |
|---|---|
| kind | Peer-reviewed paper |
| authors | Jianfeng Xiang, Zelong Lv, Sicheng Xu, Yu Deng, Ruicheng Wang, Bowen Zhang, Dong Chen, Xin Tong, Jiaolong Yang |
| venue | CVPR 2025 |
| date | 2024-12 |
| url | https://github.com/microsoft/TRELLIS |
| arxiv | 2412.01506 |
| accessed | 2026-10-24 |

## Notes

TRELLIS: Structured LATent (sparse voxels with vision-foundation-model features); up to 2B parameters; ~500K 3D objects; decodes to radiance fields, 3D Gaussians, meshes; repo lists MIT license, 16 GB GPU, CVPR'25 spotlight.

## Backlinks (computed)

**cited by:** [Text- and image-to-3D asset generation](../technique/text-to-3d-assets.md)

## Citation

Jianfeng Xiang, Zelong Lv, Sicheng Xu, Yu Deng, Ruicheng Wang, Bowen Zhang, Dong Chen, Xin Tong, Jiaolong Yang (2024-12). *Structured 3D Latents for Scalable and Versatile 3D Generation*. CVPR 2025 https://github.com/microsoft/TRELLIS

---
Provenance: created in run `2026-10-24-techniques-world-models`.