# allenai/objaverse-xl

🪐 Objaverse-XL is a Universe of 10M+ 3D Objects. Contains API Scripts for Downloading and Processing!

Repository: https://github.com/allenai/objaverse-xl
Canonical: https://ross.abutalabs.com/products/objaverse-xl
Homepage: https://objaverse.allenai.org/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: 3d, objaverse, python, blender, computer-vision, image-to-3d, text-to-3d
Last push: 2024-08-27T17:45:06+00:00

## Health v2 (maintenance only)
Score: 18/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 77
- inputs: {"age_days": 1083, "days_push": 736, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1351, forks 90 (observed 2026-08-28T04:04:28.168635+00:00)

## What it is
Objaverse-XL is an open dataset of over 10 million annotated 3D objects, along with Python API scripts for downloading, processing, and Blender-rendering the assets. It was used to train Zero123-XL, a foundation model for 3D generation supporting image-to-3D and text-to-3D workflows.

## Use cases
- download a large 3D object dataset for machine learning
- render 3D objects with Blender for training data
- train 3D foundation models like Zero123-XL
- generate 3D models from a single image
- generate 3D assets from text prompts
- research novel view synthesis on sketches and cartoons

## When to choose
- you need a massive, diverse corpus of 3D objects for training or evaluating 3D models
- you want a Python API to programmatically download and process 3D assets
- you need Blender-compatible rendering pipelines for 3D data

## When to avoid
- you need a small curated dataset with consistent licensing for commercial use
- you only need lightweight 3D assets for a game or app rather than ML research
- you cannot handle multi-terabyte downloads or GPU-heavy processing

## Facets
- artifact type: dataset
- maturity: stable
- function: machine-learning, computer-vision, data-science, etl
- domain: computer-vision, machine-learning, artificial-intelligence, graphics
- platform: python, cross-platform
- tags: 3d-objects, 3d-dataset, image-to-3d, text-to-3d, blender-rendering, novel-view-synthesis, huggingface, zero123-xl

## Member repositories
- allenai/objaverse-xl (main) score 18

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:28.168635+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:42:12.545230+00:00, confidence not recorded.
  - readme: https://github.com/allenai/objaverse-xl (fetched 2026-08-28T04:04:28.168635+00:00, sha e25535dee562)
  - homepage: https://objaverse.allenai.org/ (fetched 2026-08-29T12:00:52.106693+00:00, sha 64db10e69b4c)
  - site_page: https://objaverse.allenai.org/docs/intro (fetched 2026-08-29T12:00:52.109371+00:00, sha 7585fa684c52)
- Data as of 2026-08-30T08:39:29.467469+00:00.