🚀 Convert any image into structured, editable 3D meshes with precise geometry, topology control, texturing, and GLB export.
▶️ Run Directly in Cloud:
https://www.runcomfy.com/comfyui-workflows/comfyui-trellis2-workflow-advanced-structure-image-generation?utm_source=civitai
💡 Overview
ComfyUI Trellis2 transforms input images into structured voxel-based 3D meshes with strong geometric consistency. Instead of generating flat images, Trellis2 produces editable 3D geometry that preserves the visual structure of the source image.
Designed for 3D artists, technical artists, and designers — it provides both simple and advanced image-to-mesh pipelines with full control over mesh resolution, voxel density, and generation behavior.
✨ Key Features
Image-to-3D Pipeline: Convert photos or concept art into structured 3D meshes in one pass.
Mesh Refinement: Dedicated stages for remeshing, simplification, and topology optimization.
UV & Texturing: Full post-processing with UV unwrapping and rasterized texture generation for production-ready outputs.
GLB Export: Clean exports ready for real-time previews, game engines, or external 3D tools.
Multi-View Support: Accepts multi-view image inputs for enhanced geometric accuracy.
🚀 Getting Started
Upload your source image — a photo, concept art, or rendered object.
Choose your pipeline — simple (fast) or advanced (full control over resolution and voxel density).
Generate — the workflow produces a textured 3D mesh and exports a GLB file.
Click the "Run Directly" link above to bypass local setup and test this workflow immediately in your browser.
Description
Initial release.
FAQ
Comments (1)
It looks really interesting. But, can you please share some technical information. For example: what is the size of the required models, maybe links to used models, how much VRAM and RAM is used for generation? For example, this horse: how much memory was used when generating it? How long did it take? Will this work on 8 GB of VRAM or 16 GB?