# yashbhalgat/HashNeRF-pytorch

Pure PyTorch Implementation of NVIDIA paper on Instant Training of Neural Graphics primitives: https://nvlabs.github.io/instant-ngp/

Repository: https://github.com/yashbhalgat/HashNeRF-pytorch
Canonical: https://ross.abutalabs.com/products/hashnerf-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: nerf, real-time-rendering, computer-graphics, computer-vision, neural-network, signed-distance-functions, artificial-intelligence, machine-learning, 3d-reconstruction, efficient-training, hashing
Last push: 2024-08-07T09:25:12+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1683, "days_push": 756, "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 1037, forks 108 (observed 2026-08-28T04:03:19.400292+00:00)

## What it is
A pure PyTorch implementation of NVIDIA's Instant-NGP multi-resolution hash encoding for training NeRFs, built on top of NeRF-pytorch. It enables researchers to experiment with and extend the hash-encoding method without the original C++/CUDA codebase.

## Use cases
- train a NeRF model quickly in pure PyTorch
- experiment with multi-resolution hash encoding for neural graphics primitives
- reconstruct 3D scenes from images with NeRF
- research faster NeRF training methods
- train NeRFs on ScanNet scenes
- render novel views of synthetic objects like chairs and ficus

## When to choose
- you want a readable, hackable PyTorch codebase to modify or extend Instant-NGP's hash encoding
- you need NeRF training that runs on modest consumer GPUs
- you want to prototype NeRF research ideas without touching CUDA

## When to avoid
- you need maximum training speed or production-grade performance - use NVIDIA's original instant-ngp with tiny-cuda-nn
- you need voxel pruning or accelerated ray tracing, which are listed as TODO
- you need a maintained production library rather than a research codebase

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, graphics, simulation
- domain: computer-vision, machine-learning, graphics, artificial-intelligence
- platform: python
- tags: nerf, instant-ngp, hash-encoding, pytorch, 3d-reconstruction, neural-rendering, research, gpu, linux

## Member repositories
- yashbhalgat/HashNeRF-pytorch (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:19.400292+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 2026-08-30T07:04:25.002154+00:00, confidence not recorded.
  - readme: https://github.com/yashbhalgat/HashNeRF-pytorch (fetched 2026-08-28T04:03:19.400292+00:00, sha 4f921c0d5b8c)
- Data as of 2026-08-30T08:39:29.467469+00:00.