Papers › NVAE: A Deep Hierarchical Variational Autoencoder

NVAE: A Deep Hierarchical Variational Autoencoder

8 Jul 2020NeurIPS 2020 12arXiv:2007.03898archive 2025-07-28

Arash Vahdat, Jan Kautz

Normalizing flows, autoregressive models, variational autoencoders (VAEs), and deep energy-based models are among competing likelihood-based frameworks for deep generative learning. Among them, VAEs have the advantage of fast and tractable sampling and easy-to-access encoding networks. However, they are currently outperformed by other models such as normalizing flows and autoregressive models. While the majority of the research in VAEs is focused on the statistical challenges, we explore the orthogonal direction of carefully designing neural architectures for hierarchical VAEs. We propose Nouveau VAE (NVAE), a deep hierarchical VAE built for image generation using depth-wise separable convolutions and batch normalization. NVAE is equipped with a residual parameterization of Normal distributions and its training is stabilized by spectral regularization. We show that NVAE achieves state-of-the-art results among non-autoregressive likelihood-based models on the MNIST, CIFAR-10, CelebA 64, and CelebA HQ datasets and it provides a strong baseline on FFHQ. For example, on CIFAR-10, NVAE pushes the state-of-the-art from 2.98 to 2.91 bits per dimension, and it produces high-quality images on CelebA HQ. To the best of our knowledge, NVAE is the first successful VAE applied to natural images as large as 256×256 pixels. The source code is available at https://github.com/NVlabs/NVAE .

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 41 code samples harvested from 3 repositories linked to this paper: 26 ran (of which 15 constructed an object rather than computing a result; 21 with no instrument failure: 3 honoured, 0 violated, 18 with no contract checked; 5 where Syntology's instrument failed) · 15 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 23 samples from 1 repository, 9 ran (6 with no instrument failure; 3 where Syntology's instrument failed); community (archive-listed): 18 samples from 2 repositories, 17 ran (15 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 3 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

NVlabs/NVAE officialmentioned in papermentioned on GitHubpytorchSyntology: 23 samples, 9 ran (3 instrument-failed) · 14 unverified report
Aiwizo/template-nvae mentioned on GitHubpytorchSyntology: not harvested report
NVlabs/LSGM mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Score-based Generative Modeling in Latent Space (arXiv:2106.05931; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
NVlabs/VAEBM mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models (arXiv:2010.00654; that paper's own run record: 6 ran (of which 2 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
SerezD/NVAE-from-scratch mentioned on GitHubpytorchSyntology: 15 samples, 14 ran (2 instrument-failed) · 1 unverified report
chethankodase/alma mentioned on GitHubpytorchSyntology: not harvested report
etotheipi/nvae_tensorflow mentioned on GitHubtfUnlicenseSyntology: not harvested report
lagergren-lab/miselbo mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Multiple Importance Sampling ELBO and Deep Ensembles of Variational Approximations (arXiv:2202.10951; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
oadonca/ANVAE mentioned on GitHubtfMITSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
tcl9876/visual-vae mentioned on GitHubjaxSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

41 samples harvested: 26 ran (of which 15 constructed an object rather than computing a result; 21 with no instrument failure: 3 honoured, 0 violated, 18 with no contract checked; 5 where Syntology's instrument failed) · 15 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
3ran · our draft was wrong
2ran; the contract check could not run
18ran
15unverified

Licence: 23 of the 41 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 3 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 10 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

NVlabs/NVAE official repository 23 samples, 9 ran (3 instrument-failed) · 14 unverified
DiscMixLogistic model.py ran licence not identified · pointer only · 3b69906c3f162603 · report fixture: returned DiscMixLogistic · level recorded 2026-09-22
SE in 13 papers model.py ran · metamorphic tier: deterministic licence not identified · pointer only · 47f8add81ef809e7 · report fixture: returned SE · level recorded 2026-09-22
SyncBatchNormSwish model.py ran · metamorphic tier: deterministic licence not identified · pointer only · b014ff059854d82e · report fixture: returned SyncBatchNormSwish · level recorded 2026-09-22
get_input_size in 5 papers model.py ran · honoured contract licence not identified · pointer only · 4ca5ffe947c454b7 · report fixture: signature dataset:str/0/str · returned int · level recorded 2026-09-22
get_stride_for_cell_type model.py ran · honoured contract licence not identified · pointer only · 417c413f2e89d954 · report fixture: signature cell_type:str/0/str · returned int · level recorded 2026-09-22
groups_per_scale model.py ran · honoured contract fingerprinted licence not identified · pointer only · ec8681cb9635d8c2 · report fixture: signature num_scales:int/0/int, num_groups_per_scale:int/0/int, is_adaptive:bool/0/bool, divider:int/0/int, minimum_groups:int/0/int · returned list · level recorded 2026-09-22
mix_log_cdf_flow model.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence not identified · pointer only · 9d358b42347bc9fc · report fixture: signature z1:Tensor/2/float, logit_pi:Tensor/3/float, mu:Tensor/3/float, log_s:Tensor/3/float, log_a:Tensor/3/float, b:Tensor/3/float · returned tuple · level recorded 2026-09-22
normalize_weight_jit model.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant licence not identified · pointer only · 509ea43acade44e9 · report fixture: returned Tensor · level recorded 2026-09-22
one_hot in 150 papers
census: 18 implementations from 25 papers, 10 distinct outputs across 6 buckets, one shared input per bucket this sample: bucket 2 of 6, arg 1: rank 2, kind int, dtype int64
model.py ran · our draft was wrong licence not identified · pointer only · b95bc8179cca09d6 · report fixture: returned Tensor · level recorded 2026-09-22
ARConv2d model.py unverified licence not identified · pointer only · 57dee14e4512c612 · report
ARInvertedResidual model.py unverified licence not identified · pointer only · 213dfd5e4d31949b · report
AutoEncoder in 11 papers model.py unverified licence not identified · pointer only · ed8aab6b16f71a33 · report
Cell in 12 papers model.py unverified licence not identified · pointer only · 095d98d4a1f55bfd · report
CellAR model.py unverified licence not identified · pointer only · a0ce4d356a170530 · report
Conv2D in 10 papers model.py unverified licence not identified · pointer only · a674092cc931b81d · report
DecCombinerCell model.py unverified licence not identified · pointer only · 05cbbc54fd34ad83 · report
EncCombinerCell model.py unverified licence not identified · pointer only · 464aa3f0a27bcf6b · report
MixLogCDFParam model.py unverified licence not identified · pointer only · c4d8db6435b9c07a · report
Normal in 11 papers model.py unverified licence not identified · pointer only · 277ed0e40f07a938 · report
NormalDecoder model.py unverified licence not identified · pointer only · da4e99992612094c · report
PairedCellAR model.py unverified licence not identified · pointer only · 04bf8ed49665f416 · report
average_tensor model.py unverified licence not identified · pointer only · dd83a8a0b04c47fd · report
get_skip_connection model.py unverified licence not identified · pointer only · b8e68f7d505c301c · report
SerezD/NVAE-from-scratch community (archive-listed) 15 samples, 14 ran (2 instrument-failed) · 1 unverified
DecCombinerCell src/model.py ran fingerprinted MIT (permissive) · e79daa47e7ea4581 · report fixture: returned DecCombinerCell · level recorded 2026-09-22
DiscMixLogistic src/model.py ran MIT (permissive) · dc14f24bd29f3b0d · report fixture: returned DiscMixLogistic · level recorded 2026-09-22
EncCombinerCell src/model.py ran fingerprinted MIT (permissive) · fd98ee5b850457e9 · report fixture: returned EncCombinerCell · level recorded 2026-09-22
MaskedConv2d in 7 papers src/model.py ran fingerprinted MIT (permissive) · ea5de3e39f202a06 · report fixture: returned MaskedConv2d · level recorded 2026-09-22
NFBlock src/model.py ran fingerprinted MIT (permissive) · 2bf1bbe0e0e1f51c · report fixture: returned NFBlock · level recorded 2026-09-22
NFCell src/model.py ran fingerprinted MIT (permissive) · a82d266ee79c8089 · report fixture: returned NFCell · level recorded 2026-09-22
Normal in 11 papers src/model.py ran MIT (permissive) · c9ab03bfda9403fb · report fixture: returned Normal · level recorded 2026-09-22
ResidualCellDecoder src/model.py ran fingerprinted MIT (permissive) · 21cbfdf47981afd2 · report fixture: returned ResidualCellDecoder · level recorded 2026-09-22
ResidualCellEncoder src/model.py ran fingerprinted MIT (permissive) · 1d1460d11268fcbe · report fixture: returned ResidualCellEncoder · level recorded 2026-09-22
SE in 13 papers src/model.py ran fingerprinted MIT (permissive) · 852119a74dcd01db · report fixture: returned SE · level recorded 2026-09-22
SkipDown src/model.py ran fingerprinted MIT (permissive) · 90b219a8fe64989e · report fixture: returned SkipDown · level recorded 2026-09-22
SkipUp src/model.py ran fingerprinted MIT (permissive) · a8aa4571eb4d2e83 · report fixture: returned SkipUp · level recorded 2026-09-22
gumbel_sampling in 5 papers src/model.py ran · our draft was wrong fingerprinted MIT (permissive) · cb77f2d04693965c · report fixture: signature logits:Tensor/2/float, temperature:float/0/float · returned Tensor · level recorded 2026-09-22
soft_clamp in 14 papers src/model.py ran · our draft was wrong fingerprinted MIT (permissive) · 081e9e26fd1942f7 · report fixture: signature x:Tensor/2/float, n:float/0/float · returned Tensor · level recorded 2026-09-22
AutoEncoder in 11 papers src/model.py unverified MIT (permissive) · 2b09442d4473a79b · report
oadonca/ANVAE community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
Global_Average_Pooling modules.py ran MIT (permissive) · 5f0c5cf7f773ac3b · report fixture: signature x:EagerTensor/4/float · returned EagerTensor · level recorded 2026-09-24
batchNorm modules.py ran MIT (permissive) · c5a084857e4bb9e0 · report fixture: signature layer_dict:dict/0/?, name:str/0/str · returned SymbolicTensor · level recorded 2026-09-24
swish in 258 papers
census: 14 implementations from 37 papers, 2 distinct outputs on one shared input
modules.py ran MIT (permissive) · 248788c293d8eccd · report fixture: signature layer_dict:dict/0/?, name:str/0/str · returned EagerTensor · level recorded 2026-09-24

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for NVAE (graph method: NVAE). 15 bricks; individually on a fixture: 14 ran (of which 12 constructed an object rather than computing a result; 12 with no instrument failure: 0 honoured, 0 violated, 12 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

  1. AutoEncoder entry · level 0 · unverified · 2b09442d4473a79b
  2. DecCombinerCell model · level 1 · ran · e79daa47e7ea4581
  3. DiscMixLogistic model · level 1 · ran · dc14f24bd29f3b0d
  4. EncCombinerCell model · level 1 · ran · fd98ee5b850457e9
  5. MaskedConv2d helper · level 1 · ran · ea5de3e39f202a06
  6. NFBlock model · level 1 · ran · 2bf1bbe0e0e1f51c
  7. Normal model · level 1 · ran · c9ab03bfda9403fb
  8. ResidualCellDecoder model · level 1 · ran · 21cbfdf47981afd2
  9. ResidualCellEncoder model · level 1 · ran · 1d1460d11268fcbe
  10. gumbel_sampling helper · level 1 · ran · our draft was wrong · cb77f2d04693965c
  11. NFCell model · level 1 · ran · a82d266ee79c8089
  12. soft_clamp helper · level 1 · ran · our draft was wrong · 081e9e26fd1942f7
  13. SE helper · level 1 · ran · 852119a74dcd01db
  14. SkipUp helper · level 1 · ran · a8aa4571eb4d2e83
  15. SkipDown helper · level 1 · ran · 90b219a8fe64989e

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2007.03898") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image Generation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Generation CIFAR-10 NVAE w/ flow FID 32.53 #70 of 78 Archive leaderboard report
Image Generation CelebA 256x256 NVAE w/ flow bpd 0.70 #5 of 17 Archive leaderboard report
Image Generation FFHQ 256 x 256 NVAE w/ flow bits/dimension 0.69 #51 of 51 Archive leaderboard report
Image Generation ImageNet 32x32 NVAE w/ flow bpd 3.92 #29 of 35 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: NVAE Encoder Residual Cell, NVAE Generative Residual Cell, Residual Normal Distribution

1x1 ConvolutionAdaMaxAverage PoolingBatch NormalizationConvolutionCosine AnnealingDense ConnectionsDepthwise ConvolutionDepthwise Separable ConvolutionHierarchical VAENVAENVAE Encoder Residual CellNVAE Generative Residual CellNormalizing FlowsPointwise ConvolutionReLUResidual ConnectionResidual Normal DistributionSigmoid ActivationSpectral NormalizationSqueeze-and-Excitation BlockWeight Normalization

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