Fast3D generates 3D models from text prompts or images, with material and texture generation.
Free trial
9 alternativesTrellis
Microsoft TRELLIS is a research repository for generating structured 3D assets from images or text, according to its project topics.
About Trellis
Microsoft TRELLIS is an open-source research repository for generating structured 3D assets from text or image prompts. It uses a unified Structured LATent (SLAT) representation and Rectified Flow Transformers as backbones, with pre-trained models up to 2 billion parameters trained on a 500K diverse 3D object dataset. It can output generated assets in multiple formats including Radiance Fields, 3D Gaussians, and meshes, and supports basic 3D editing like variant generation and local asset editing. The repository is only tested on Linux systems, requires an NVIDIA GPU with at least 16GB of memory, CUDA Toolkit 11.8 or 12.2, and Python 3.8 or higher. Text-conditioned generation models are noted to be less creative and detailed than image-conditioned models due to data limitations, and the tuning-free multi-image conditioning feature may not produce optimal results for all input images.
Key features
- Generates diverse 3D assets with detailed shape and texture from text or image prompts
- Outputs 3D assets in Radiance Fields, 3D Gaussians, and mesh formats
- Supports generation of asset variants from existing 3D models
- Allows local editing of generated 3D assets
- Includes pre-trained models up to 2 billion parameters
- Provides training code and dataset toolkits for 3D asset preparation
Best for: Researchers, 3D content creators, and developers who need to generate or edit structured 3D assets from text or image prompts.
Good for
- Supports multiple 3D output formats to accommodate diverse downstream use cases
- Includes flexible 3D editing capabilities like variant generation and local editing
- Open-source with access to pre-trained models, training code, and dataset toolkits
Watch out for
- Text-conditioned models are less creative and detailed than image-conditioned models due to data limitations
- Tuning-free multi-image conditioning may not produce optimal results for all input images
- Code is only tested on Linux, with no fully tested Windows support
Trellis at a glance
- Website
- github.com/microsoft/TRELLIS
- Category
- Design & Creative › 3D, CAD & Motion
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
Trellis alternatives
Tools and sites you could use instead, ranked by how closely they match.
ThreeDee generates 3D models from text descriptions or images using an AI model.
YellowSculpt generates clean-topology 3D character meshes from prompts, rigged and ready for animation.
Dream Textures integrates Stable Diffusion into Blender, according to its repository description, with image-generation topics listed.
Open source
18 alternatives
This Blender add-on connects the RenderMan renderer with Blender, according to its repository description.
Open source
18 alternatives
The MCprep repository describes a Blender Python add-on for Minecraft renders and animations, intended to improve that workflow.
Open source
20 alternativesPopular in 3D, CAD & Motion
Well-known listings from the same topic.
NASA 3D Resourcesscience.nasa.gov/3d-resourcesDesignConnecteddesignconnected.comAwesome Blenderawesome-blender.netlify.appWavelrywavelry.vercel.appJerrari Designjerrari3d.comAssemblrassemblrworld.com
Trellis FAQ
What input types does TRELLIS support for 3D generation?
It accepts both text and image prompts as inputs for 3D asset generation.
What 3D output formats can TRELLIS generate?
It can output generated assets as Radiance Fields, 3D Gaussians, and meshes, among other formats.
What are the minimum system requirements to run TRELLIS?
It requires a Linux system, an NVIDIA GPU with at least 16GB of memory, CUDA Toolkit 11.8 or 12.2, and Python 3.8 or higher.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.