# NVIDIAGameWorks/kaolin-wisp

NVIDIA Kaolin Wisp is a PyTorch library powered by NVIDIA Kaolin Core to work with neural fields (including NeRFs, NGLOD, instant-ngp and VQAD).

Repository: https://github.com/NVIDIAGameWorks/kaolin-wisp
Canonical: https://ross.abutalabs.com/products/kaolin-wisp
Language: Python
License: NOASSERTION
License Family: other
Topics: 3d-deep-learning, nerf, pytorch, sdf, neuralfields
Last push: 2024-08-04T20:36:42+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 1499, "days_push": 759, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1498, forks 141 (observed 2026-08-28T04:04:54.034566+00:00)

## What it is
NVIDIA Kaolin Wisp is a PyTorch library and engine for neural fields research, built on NVIDIA Kaolin Core. It provides differentiable renderers, data structures (octrees, hash grids, triplanar features), datasets, interactive visualization, and trainer classes for models like NeRFs, NGLOD, instant-ngp, and VQAD.

## Use cases
- train NeRF models from images
- research neural signed distance fields
- build custom neural field architectures with hash grids and octrees
- interactively visualize and debug neural field training
- render novel views of 3D scenes
- experiment with instant-ngp style multiresolution hash grids

## When to choose
- you need a flexible PyTorch framework for neural fields research
- you want differentiable rendering and sparse data structures out of the box
- you need interactive training visualization on NVIDIA GPUs
- you want reference implementations of NeRF, NGLOD, and instant-ngp variants

## When to avoid
- you need a production-ready 3D reconstruction pipeline rather than a research toolkit
- you don't have a CUDA-capable NVIDIA GPU
- you need a simple one-command NeRF trainer with minimal configuration
- your project requires a permissive open-source license (NVIDIA Source Code License applies)

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, graphics, simulation, image-processing, data-visualization
- domain: deep-learning, machine-learning, graphics, computer-vision, developer-tools
- platform: python, windows
- tags: neural-fields, nerf, sdf, pytorch, 3d-reconstruction, differentiable-rendering, nvidia, research, gpu, linux

## Member repositories
- NVIDIAGameWorks/kaolin-wisp (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:54.034566+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 2026-08-30T04:33:04.047305+00:00, confidence not recorded.
  - readme: https://github.com/NVIDIAGameWorks/kaolin-wisp (fetched 2026-08-28T04:04:54.034566+00:00, sha 0976723d31ae)
- Data as of 2026-08-30T08:39:29.467469+00:00.