# google/GNM

An open ecosystem of parametric human models and perception stacks, starting with GNM Head.

Repository: https://github.com/google/GNM
Canonical: https://ross.abutalabs.com/products/gnm
Language: Python
License: Apache-2.0
License Family: permissive
Topics: 3dmm, gnm, parametric-model, digital-human
Last push: 2026-08-25T18:09:09+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 11
- inputs: {"age_days": 163, "days_push": 8, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1469, forks 168 (observed 2026-08-28T04:04:49.065195+00:00)

## What it is
GNM is an open ecosystem of parametric statistical human models and perception stacks from Google, starting with GNM Head, a high-fidelity 3D morphable model of the human head with controllable identity, expression, pose, and internal anatomy. It provides multi-framework backends for NumPy, JAX, PyTorch, and TensorFlow under an Apache 2.0 license.

## Use cases
- generate 3D human head meshes with controllable identity and expression
- fit a 3D morphable model to face images or scans
- sample statistically plausible head shapes for synthetic data generation
- model eyeballs, teeth, and tongue anatomy in 3D avatars
- build face perception or reconstruction pipelines in PyTorch or JAX
- create digital humans for graphics and generative AI research

## When to choose
- you need a state-of-the-art, openly licensed 3DMM of the human head
- you want multi-framework support (NumPy, JAX, PyTorch, TensorFlow)
- you need fine-grained disentangled control over identity, expression, and pose
- you require internal anatomy like eyeballs, teeth, and tongue in head models

## When to avoid
- you need full-body parametric human models rather than heads
- you need a ready-made application or GUI rather than a modeling library
- your project depends on models not yet released in the ecosystem roadmap

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, graphics, simulation, sdk
- domain: computer-vision, graphics, artificial-intelligence, machine-learning
- platform: python, cross-platform
- tags: 3dmm, parametric-model, digital-human, 3d-face, human-head-model, jax, pytorch, tensorflow, numpy

## Member repositories
- google/GNM (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:49.065195+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 2026-08-30T04:34:52.996880+00:00, confidence not recorded.
  - readme: https://github.com/google/GNM (fetched 2026-08-28T04:04:49.065195+00:00, sha c450ff029fe3)
- Data as of 2026-08-30T08:39:29.467469+00:00.