TRELLIS
Microsoft's MIT-licensed 3D foundation model: text or image in, textured mesh out.
About
TRELLIS is Microsoft's unified 3D generation model, published December 2024 (arXiv 2412.01506). It uses structured latent representations (SLAT) decoded to both explicit meshes via FlexiCubes and NeRF-style radiance fields, generating textured, part-aware 3D assets from a single image or a text prompt.
Models and most of the code are MIT licensed, making it one of the most commercially friendly 3D generators available, with the 12B-class pipeline downloadable from the official GitHub. It is heavyweight — around 12GB of weights and CUDA-centric tooling — and outputs are static meshes, so animation rigging is a separate step.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (metrics tier, metrics-v1). Evidence: Microsoft Research release; CVPR 2025 Spotlight; successors (v2, v3) followed; MIT license on models and most code; Heavyweight (~12GB, CUDA-centric); static meshes, no animation rigging. License: MIT.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 7.0/10
- Openness · license + docs
- 10.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · Microsoft Research release; CVPR 2025 Spotlight; successors (v2, v3) followed
- · MIT license on models and most code
- · Heavyweight (~12GB, CUDA-centric); static meshes, no animation rigging