# microsoft/TRELLIS.2

Native and Compact Structured Latents for 3D Generation

Repository: https://github.com/microsoft/TRELLIS.2
Canonical: https://ross.abutalabs.com/products/trellis2
Language: Python
License: MIT
License Family: permissive
Last push: 2026-07-10T12:34:32+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 91, release rhythm 35, longevity 20
- inputs: {"age_days": 280, "days_push": 54, "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 10869, forks 1325 (observed 2026-08-28T04:10:44.541206+00:00)

## What it is
TRELLIS.2 is a 4B-parameter state-of-the-art 3D generative model from Microsoft for high-fidelity image-to-3D asset generation. It uses a field-free sparse voxel representation (O-Voxel) and a sparse 3D VAE to produce meshes with complex topology and full PBR materials.

## Use cases
- generate 3d models from a single image
- create textured 3d assets with pbr materials
- reconstruct meshes with open surfaces and non-manifold geometry
- convert textured meshes to compact voxel latents
- generate 3d assets for games and rendering pipelines

## When to choose
- you need high-resolution image-to-3D generation with PBR textures
- your assets have complex topology like open surfaces or internal structures
- you have a high-memory NVIDIA GPU (24GB+) and Linux environment

## When to avoid
- you need a lightweight model that runs on consumer GPUs or CPUs
- you require Windows or macOS support
- you only need simple low-poly 3D generation

## 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, image-to-3d, generative-model, sparse-voxel, pbr-textures, 3d-assets, diffusion-transformer, huggingface, game-development, linux, gpu, docker

## Member repositories
- microsoft/TRELLIS.2 (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:44.541206+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 2026-08-29T17:17:27.556528+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/TRELLIS.2 (fetched 2026-08-28T04:10:44.541206+00:00, sha 9c99abda22ac)
- Data as of 2026-08-30T08:39:29.467469+00:00.