objaverse

What it is and what it does

Objaverse is a companion package for accessing and processing Objaverse-XL, an open dataset containing over 10 million 3D objects. The package provides utilities to download and work with this large-scale 3D model collection, which was created to support training of 3D foundation models like Zero123-XL. It depends on requests, pandas, pyarrow, tqdm, loguru, fsspec, and gputil to handle data retrieval, processing, progress tracking, and resource monitoring across potentially large downloads.

The package is designed for researchers and developers working with 3D generative models, novel view synthesis, and 3D reconstruction tasks. It integrates with Hugging Face for dataset hosting and includes support for Blender rendering workflows. Note that the project is dormant—the last release was in November 2023—so it receives no active maintenance or updates.

Use it for:

• Download subsets of the Objaverse-XL dataset for training 3D vision models or foundation models.
• Batch process 3D objects for rendering, format conversion, or metadata extraction in machine learning pipelines.
• Access pre-processed 3D model collections for research in novel view synthesis or 3D reconstruction.
• Integrate 3D object data into computer vision projects that require diverse, large-scale 3D training data.
• Retrieve and organize 3D models for use with Blender rendering scripts or other 3D graphics workflows.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Downloads and processes Objaverse-XL, an open dataset of over 10 million 3D objects, with utilities for managing large-scale 3D model data retrieval and preparation.

Yes, if you are actively working with Objaverse-XL for 3D model research or training. The package has low install friction and permissive licensing. However, be aware that maintenance is dormant (last update November 2023), so expect no bug fixes or compatibility updates. Verify that the dataset access and your intended use case align with the ODC-By v1.0 license and any per-object restrictions before committing to a large download.

Install

objaverse on PyPI

pip install objaverse

uv add objaverse

poetry add objaverse

Installing objaverse

Before you install

Low install friction with a pure-Python wheel and standard data-handling dependencies. Maintenance is dormant (last release 2023-11-01, 1017 days ago), so expect no active bug fixes or updates.

License in practice

Licensed under Apache Software License (permissive), meaning you can use, modify, and distribute the package freely provided you include the license notice. The underlying dataset itself uses ODC-By v1.0 with per-object license variation.

Quickstart

pip install objaverse

import objaverse
# Use the package to download and process 3D objects from Objaverse-XL

Polycam data requires separate academic approval; large downloads may require significant disk space and bandwidth.

Verify before relying

• Exact Python version compatibility (requires_python is unspecified in the fact sheet)
• Whether the package provides a high-level download API or primarily serves as a companion to external scripts
• Current state of the dataset's availability and any access restrictions beyond Polycam data

Package facts

Evidence: objaverse-0.1.7-py3-none-any.whl

Tags

More Artificial Intelligence packages