# 3DTopia/3DTopia-XL

[CVPR 2025 Highlight] 3DTopia-XL: High-Quality 3D PBR Asset Generation via Primitive Diffusion

Repository: https://github.com/3DTopia/3DTopia-XL
Canonical: https://ross.abutalabs.com/products/3dtopia-xl
Homepage: https://3dtopia.github.io/3DTopia-XL/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: 3d-generation, diffusion-models, image-to-3d, text-to-3d, cvpr, cvpr2025
Last push: 2025-07-14T01:54:53+00:00

## Health v2 (maintenance only)
Score: 37/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 31, release rhythm 35, longevity 54
- inputs: {"age_days": 763, "days_push": 416, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1045, forks 40 (observed 2026-08-28T04:03:21.535985+00:00)

## What it is
3DTopia-XL is a 3D diffusion transformer model that generates high-quality 3D assets with PBR materials from a single image or text prompt in about five seconds. It uses a primitive-based 3D representation (PrimX) and exports GLB meshes ready for graphics pipelines like Blender.

## Use cases
- generate 3d models from a single image
- text to 3d asset generation
- create pbr textured meshes for game engines
- convert photos into 3d models
- generate 3d assets for blender rendering
- research on 3d diffusion transformers

## When to choose
- you need fast generation of game-ready 3D assets with PBR materials from images or text
- you want to experiment with or build on state-of-the-art 3D generative diffusion research
- you need GLB exports that integrate into standard graphics pipelines

## When to avoid
- you need precise, artist-controlled 3D modeling rather than generative output
- you lack a CUDA GPU, since inference and training require significant GPU resources
- you need production-grade asset pipelines with guarantees on topology or mesh quality

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, graphics, simulation
- domain: artificial-intelligence, deep-learning, computer-vision, graphics
- platform: python
- tags: 3d-generation, diffusion-transformer, text-to-3d, image-to-3d, pbr-materials, primx, asset-generation, cvpr-2025, game-development, gpu, linux

## Member repositories
- 3DTopia/3DTopia-XL (main) score 37

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:21.535985+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:02:01.600791+00:00, confidence not recorded.
  - readme: https://github.com/3DTopia/3DTopia-XL (fetched 2026-08-28T04:03:21.535985+00:00, sha fab940c8d232)
  - homepage: https://3dtopia.github.io/3DTopia-XL/ (fetched 2026-08-29T13:03:05.765577+00:00, sha 65acca7cdd29)
- Data as of 2026-08-30T08:39:29.467469+00:00.