Papers › Instant Neural Graphics Primitives with a Multiresolution Hash Encoding

Instant Neural Graphics Primitives with a Multiresolution Hash Encoding

16 Jan 2022arXiv:2201.05989archive 2025-07-28

Thomas Müller, Alex Evans, Christoph Schied, Alexander Keller

Neural graphics primitives, parameterized by fully connected neural networks, can be costly to train and evaluate. We reduce this cost with a versatile new input encoding that permits the use of a smaller network without sacrificing quality, thus significantly reducing the number of floating point and memory access operations: a small neural network is augmented by a multiresolution hash table of trainable feature vectors whose values are optimized through stochastic gradient descent. The multiresolution structure allows the network to disambiguate hash collisions, making for a simple architecture that is trivial to parallelize on modern GPUs. We leverage this parallelism by implementing the whole system using fully-fused CUDA kernels with a focus on minimizing wasted bandwidth and compute operations. We achieve a combined speedup of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds, and rendering in tens of milliseconds at a resolution of 1920×1080.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2201.05989")

Code

Syntology Ran 18 of 44 code samples harvested from 11 repositories linked to this paper; 26 have no recorded run. Of those that ran: 2 ran · honoured contract; 4 ran · our draft was wrong; 5 ran · fixture could not drive it; 7 ran with no contract checked.

By repository: official repository: 7 samples from 1 repository, 2 ran; community (archive-listed): 36 samples from 10 repositories, 15 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

17 repositories listed; official and paper-mentioned ones first.

nvlabs/instant-ngp officialmentioned in papermentioned on GitHubpytorchNOASSERTION report
nvlabs/tiny-cuda-nn officialmentioned in papermentioned on GitHubpytorch report
bycloudai/instant-ngp-windows mentioned on GitHubNOASSERTION report
Jittor/JNeRF Apache-2.0 report
blurgyy/jaxngp jaxApache-2.0 report
kwea123/ngp_pl pytorchMIT report
nvidiagameworks/kaolin-wisp pytorchNOASSERTION report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

44 samples harvested; 18 ran; 2 honoured the contract we drafted; 26 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
4ran · our draft was wrong
5ran · fixture could not drive it
7ran
26unverified

Licence: 9 of the 44 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 11 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

null_tensor_like nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository ran · our draft was wrong fingerprinted licence not identified · pointer only · 358332bd1fcb1f00 · report
null_tensor_to_none nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository ran · honoured contract fingerprinted licence not identified · pointer only · dac732e0c07c3548 · report
Encoding nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository unverified licence not identified · pointer only · 1bc1df6613a61f7f · report
Module nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository unverified licence not identified · pointer only · ee42d0c97cb735ff · report
_module_function nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository unverified licence not identified · pointer only · 37829527c0540e77 · report
_module_function_backward nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository unverified licence not identified · pointer only · 2d3a269acf2a7eb9 · report
_torch_precision nvlabs/tiny-cuda-nn/bindings/torch/tinycudann/modules.py official repository unverified licence not identified · pointer only · 27104b2bbb1f407d · report
Encoding nerfstudio-project/nerfstudio/nerfstudio/field_components/encodings.py community (archive-listed) ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 5881a1312a6c8239 · report
GridEncoder ashawkey/torch-ngp/gridencoder/grid.py community (archive-listed) ran MIT (permissive) · 58ab43a58e36a71b · report
HashEncoding nerfstudio-project/nerfstudio/nerfstudio/field_components/encodings.py community (archive-listed) ran · metamorphic tier: deterministic Apache-2.0 (permissive) · e5c0d600b3cfd7df · report
LevelInfo cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran MIT (permissive) · 6838aa47a20dbb27 · report
_bilinear_params_2d cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · ecbd5905781da167 · report
_bilinear_params_3d cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · 15c83cf85bc3d3a8 · report
_compute_bilinear_params cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · baff6a03b08f0445 · report
_compute_resolutions cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran · honoured contract fingerprinted MIT (permissive) · 82b510e49ca33bd4 · report
_grid_encode ashawkey/torch-ngp/gridencoder/grid.py community (archive-listed) ran MIT (permissive) · 2a16d349fb039166 · report
_hash_ravel cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · 677c2f6def386f0c · report
_hash_xor cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 3da7a3fe9ba479ea · report
_query kair-bair/nerfacc/nerfacc/grid.py community (archive-listed) ran · fixture could not drive it fingerprinted licence not identified · pointer only · 74411ce6cbc0ec6d · report
run_network yashbhalgat/HashNeRF-pytorch/run_nerf.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 296bbd3954dea52c · report
srgb_to_linear Jittor/JNeRF/python/jnerf/dataset/dataset_util.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · e4806ed5a1eb8585 · report
FieldComponent nerfstudio-project/nerfstudio/nerfstudio/field_components/encodings.py community (archive-listed) unverified Apache-2.0 (permissive) · bab4f72831565f5b · report
MultiLevelHybridHashEncoding cheind/pure-torch-ngp/torchngp/modules/encoding.py community (archive-listed) unverified MIT (permissive) · 31c12ec24d6df685 · report
batchify_rays yashbhalgat/HashNeRF-pytorch/run_nerf.py community (archive-listed) unverified MIT (permissive) · aa4cda8b1a0496ec · report
batchify_rays JonathonLuiten/instant-nerf-pytorch/run_nerf.py community (archive-listed) unverified MIT (permissive) · 8745b56380aead90 · report
depth2img Linyou/taichi-ngp-renderer/taichi_ngp.py community (archive-listed) unverified MIT (permissive) · 6d1525352ca6a17a · report
empty_impl blurgyy/jaxngp/utils/types.py community (archive-listed) unverified Apache-2.0 (permissive) · ae252edecc5fa2f3 · report
eval_step blurgyy/jaxngp/app/imagefit.py community (archive-listed) unverified Apache-2.0 (permissive) · e1d0a3ae170c33ea · report
f32_to_u8 blurgyy/jaxngp/utils/data.py community (archive-listed) unverified Apache-2.0 (permissive) · e8a2272fdb3aab11 · report
fov_to_focal_length Jittor/JNeRF/python/jnerf/dataset/dataset_util.py community (archive-listed) unverified Apache-2.0 (permissive) · 17904ab5cfa8d0aa · report
load_K_Rt_from_P Jittor/JNeRF/python/jnerf/dataset/neus_dataset.py community (archive-listed) unverified Apache-2.0 (permissive) · 69986a1a9ebdac78 · report
make_near_far_from_bound blurgyy/jaxngp/models/renderers/cuda.py community (archive-listed) unverified Apache-2.0 (permissive) · 6ee9d4248fb29f6e · report
mkValueError blurgyy/jaxngp/utils/common.py community (archive-listed) unverified Apache-2.0 (permissive) · e3fa3c825ade63a5 · report
mono_to_rgb blurgyy/jaxngp/utils/data.py community (archive-listed) unverified Apache-2.0 (permissive) · ea5f6eb138cf8162 · report
print_tcnn_speed_warning nerfstudio-project/nerfstudio/nerfstudio/field_components/encodings.py community (archive-listed) unverified Apache-2.0 (permissive) · 64bd11eb81b5879b · report
read_image_imageio Jittor/JNeRF/python/jnerf/dataset/dataset_util.py community (archive-listed) unverified Apache-2.0 (permissive) · 30eef9632b25220c · report
render JonathonLuiten/instant-nerf-pytorch/run_nerf.py community (archive-listed) unverified MIT (permissive) · e9fd994e70bf07d4 · report
replace_impl blurgyy/jaxngp/utils/types.py community (archive-listed) unverified Apache-2.0 (permissive) · 6e4f132e09f5d5c1 · report
run_network JonathonLuiten/instant-nerf-pytorch/run_nerf.py community (archive-listed) unverified MIT (permissive) · fc1a65ccede65297 · report
to_cpu blurgyy/jaxngp/utils/data.py community (archive-listed) unverified Apache-2.0 (permissive) · 52787ea8b908e7b9 · report
train_step blurgyy/jaxngp/app/imagefit.py community (archive-listed) unverified Apache-2.0 (permissive) · b2c79c7c992d6668 · report
vmap_jaxfn_with blurgyy/jaxngp/utils/common.py community (archive-listed) unverified Apache-2.0 (permissive) · 2105a13cbfc9df70 · report
batchify identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · fd57e365830d09de · report

Tasks

3D Reconstruction3D Shape ReconstructionNeural Radiance CachingNeural RenderingNovel View Synthesis

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamDense ConnectionsExponential DecayFeedforward NetworkReLU

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections