Scal3R: 3D Generator — What It Does and What You Need to Run It

Scal3R is an open source 3D generator from the ZJU3DV group. The tool solves the image-to-3D task: an image goes in, a 3D reconstruction comes out. The authors position the project as scalable test-time training for large-scale 3D reconstruction and note that the work was accepted to CVPR 2026 with a Highlight designation.

What it does

As described by the author, Scal3R is intended for large-scale 3D reconstruction from images. The repository tags list three key areas:

• 3D reconstruction;
• depth estimation;
• large-scene handling.

The authors do not provide a detailed list of supported input image formats or output meshes in the fact sheet. Scene size limits and requirements for the number of views are not specified either.

What you need to run it

The code is written in Python. According to the author's description, CUDA is required — the repository includes a separate installation guide with instructions for PyTorch and CUDA. Specific library versions and the minimum amount of VRAM are not listed in the fact sheet.

The code is distributed under the MIT License: any use, including commercial, modification and redistribution, is permitted provided the copyright notice and license text are retained. The code is open source, and the model is published on Hugging Face.

Who it suits

The tool will be of interest to researchers and developers working on 3D scene reconstruction and depth estimation who want to explore the test-time training approach. The open source code and the published model make it possible to understand the method and adapt it to your own data.

For everyday tasks like "upload one photo — get a ready model for printing," the project may be overkill: the authors do not describe a ready-made interface, and installation requires working with CUDA and PyTorch. Those who are not ready to deal with manual environment setup should look at other solutions.

Scal3R is a research tool with open source code, aimed at large-scale reconstruction. Before using it in your project, it is worth studying the installation guide and the license terms in the repository.