Skip to content
DATASHEET // TRELLIS

TRELLIS

Microsoft Research's structured 3D representation model.

OPEN SOURCE16–24 GB VRAMRuns locally
Actually FreeNo SignupOpen SourceWatermark-Free
Visit TRELLISUPDATED 2026-04-02 · DIRECT LINK
github.com/microsoft/TRELLIS
TRELLIS — preview image

HARDWARE REQUIREMENTS //

Runs locally · High-end GPU (16–24 GB)

16–24 GB VRAM
Min VRAM
16 GB
Rec. VRAM
24 GB
Min RAM
32 GB
Rec. RAM
64 GB
Disk
30 GB
GPU class
High-end GPU
12.1+No Apple SiliconGPU RequiredQuant: FP16

Image-to-3D fits 16 GB; text-to-3D wants 24 GB.

[ EDITORIAL PICK ]

Why we recommend TRELLIS

DERIVED FROM METADATA — NOT SPONSORED
  • Open source

    Source is public — you can audit it, fork it, and you'll never lose access to your workflows if TRELLIS the company changes direction.

[ EVIDENCE NOTE ]

Documentation-led datasheet

This page summarizes upstream documentation, release information, and editorially reviewed catalogue fields. It is not presented as a hands-on benchmark. Verify changing requirements at the official project; report stale data through our corrections channel.

VRAM guide

AT-A-GLANCE SIGNALS //

DERIVED FROM THIS PAGE'S DATA
  • Install difficulty
    Standard

    A standard local install — download, install dependencies, point at your GPU.

  • Hardware comfort
    Enthusiast

    Needs 16 GB minimum — RTX 3090 / 4090 territory.

  • Ecosystem
    Open source

    Source is public — auditable and forkable, no vendor lock.

  • Verification
    Ageing

    105 days since the last catalogue refresh — flagged for re-verification.

[ COMMUNITY GUIDES & WORKFLOWS ]

Tutorials & deep-dives for TRELLIS

Hand-picked from YouTube, Reddit, GitHub, and the wider web. Each link goes straight to the source — we don't intercept or rewrite anything.

[ MORE IN THIS NICHE ]

Other heavy image generation tools we rate

Three picks across different tradeoffs — so you don't end up with three near-clones of TRELLIS.

What is TRELLIS?

TRELLIS uses a novel 'structured latent' representation — joint sparse-voxel + feature-grid — to generate high-quality 3D assets (mesh + Gaussian splat + radiance field) from images or text. MIT license, image-to-3D and text-to-3D pipelines.

Pros & cons

✓ PROS

  • Outputs mesh + Gaussian splat + radiance field in one shot
  • MIT licensed
  • Microsoft Research backing — solid engineering

– CONS

  • Heavier than Hunyuan3D-2 at equivalent quality
  • Texture quality below dedicated PBR pipelines

What's actually free?

MIT.

✓ Actually FreeNo SignupOpen SourceWatermark-Free

Alternatives

Hunyuan3D-2

Tencent's open 3D generator — multi-view, PBR, ready-to-use meshes.

OPEN SOURCE12–16 GB VRAM
VRAM fit12–16 GB

TripoSR

Single-image to 3D mesh in under a second on a 4090.

OPEN SOURCE6–8 GB VRAM
VRAM fit6–8 GB

Stable Zero123

Novel-view synthesis — generate any angle from a single image.

OPEN SOURCE6–8 GB VRAM
VRAM fit6–8 GB