# VAST-AI-Research/UniRig

[SIGGRAPH 2025] One Model to Rig Them All: Diverse Skeleton Rigging with UniRig

Repository: https://github.com/VAST-AI-Research/UniRig
Canonical: https://ross.abutalabs.com/products/unirig
Homepage: https://zjp-shadow.github.io/works/UniRig/
Language: Python
License: MIT
License Family: permissive
Topics: animation, auto-rigging, autoregressive, computer-graphics
Last push: 2026-06-04T07:15:34+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 85, release rhythm 35, longevity 36
- inputs: {"age_days": 512, "days_push": 90, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1717, forks 165 (observed 2026-08-28T04:05:26.361124+00:00)

## What it is
UniRig is the official implementation of a SIGGRAPH 2025 framework that automatically rigs 3D models by predicting skeletons and skinning weights using large autoregressive models with a bone-point cross-attention mechanism. It includes the Rig-XL dataset of over 14,000 rigged 3D models and significantly outperforms prior academic and commercial rigging methods.

## Use cases
- automatically rig a 3D character model for animation
- generate a skeleton for a 3D mesh without manual work
- predict skinning weights for anime or animal models
- speed up a 3D animation pipeline with auto-rigging
- rig diverse 3D models from AI generation or traditional workflows
- train or evaluate skeleton prediction models on the Rig-XL dataset

## When to choose
- you need automated skeleton and skinning for diverse 3D models like humans, animals, or fictional characters
- your 3D assets have complex or non-standard topologies that break simpler auto-rigging tools
- you want state-of-the-art rigging accuracy backed by a published SIGGRAPH 2025 method
- you are building an AI 3D content creation pipeline and need rigging as a step

## When to avoid
- you need fine artistic control over hand-placed bones in a tool like Blender or Maya
- you only need to rig simple humanoid characters where existing DCC auto-rig tools suffice
- you lack a GPU or cannot run large autoregressive model inference
- you need production skinning quality beyond what automated methods provide without cleanup

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, animation, graphics, llm-inference
- domain: computer-vision, graphics, machine-learning, artificial-intelligence
- platform: python
- tags: auto-rigging, 3d-graphics, skeleton-generation, skinning-weights, autoregressive-models, siggraph-2025, 3d-animation, linux, gpu

## Member repositories
- VAST-AI-Research/UniRig (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:26.361124+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 2026-08-30T03:34:12.059118+00:00, confidence not recorded.
  - readme: https://github.com/VAST-AI-Research/UniRig (fetched 2026-08-28T04:05:26.361124+00:00, sha b33efef4a9e8)
  - homepage: https://zjp-shadow.github.io/works/UniRig/ (fetched 2026-08-29T11:10:11.470873+00:00, sha ef7a0308c736)
- Data as of 2026-08-30T08:39:29.467469+00:00.