Papers › Variational Diffusion Models
Variational Diffusion Models
Diederik P. Kingma, Tim Salimans, Ben Poole, Jonathan Ho
Diffusion-based generative models have demonstrated a capacity for perceptually impressive synthesis, but can they also be great likelihood-based models? We answer this in the affirmative, and introduce a family of diffusion-based generative models that obtain state-of-the-art likelihoods on standard image density estimation benchmarks. Unlike other diffusion-based models, our method allows for efficient optimization of the noise schedule jointly with the rest of the model. We show that the variational lower bound (VLB) simplifies to a remarkably short expression in terms of the signal-to-noise ratio of the diffused data, thereby improving our theoretical understanding of this model class. Using this insight, we prove an equivalence between several models proposed in the literature. In addition, we show that the continuous-time VLB is invariant to the noise schedule, except for the signal-to-noise ratio at its endpoints. This enables us to learn a noise schedule that minimizes the variance of the resulting VLB estimator, leading to faster optimization. Combining these advances with architectural improvements, we obtain state-of-the-art likelihoods on image density estimation benchmarks, outperforming autoregressive models that have dominated these benchmarks for many years, with often significantly faster optimization. In addition, we show how to use the model as part of a bits-back compression scheme, and demonstrate lossless compression rates close to the theoretical optimum. Code is available at https://github.com/google-research/vdm .
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):
get_harvested_code_for_paper(arxiv_id="2107.00630")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="2107.00630", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2107.00630")
Code
Syntology 38 code samples harvested from 6 repositories linked to this paper: 29 ran (of which 18 constructed an object rather than computing a result; 20 with no instrument failure: 0 honoured, 1 violated, 19 with no contract checked; 9 where Syntology's instrument failed) · 9 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; 3 where the contract check could not run; failures of Syntology's instrument, not of the code.
By repository: official repository: 16 samples from 1 repository, 11 ran (7 with no instrument failure; 4 where Syntology's instrument failed); community (archive-listed): 20 samples from 4 repositories, 16 ran (12 with no instrument failure; 4 where Syntology's instrument failed); found in paper text by Syntology: 2 samples from 1 repository, 2 ran (1 with no instrument failure; 1 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.
Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.
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
38 samples harvested: 29 ran (of which 18 constructed an object rather than computing a result; 20 with no instrument failure: 0 honoured, 1 violated, 19 with no contract checked; 9 where Syntology's instrument failed) · 9 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.
Licence: 2 of the 38 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 6 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 2 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.
561db3edf2a73d49 · report
fixture: returned AttnBlock · level recorded 2026-09-22 1fd721cbc30c55b2 · report
fixture: returned Base2FourierFeatures · level recorded 2026-09-22 491caa588adc4c8f · report
fixture: returned EncDec · level recorded 2026-09-22 069d8acfc9e74bee · report
fixture: returned NoiseSchedule_FixedLinear · level recorded 2026-09-22 529196025d205b76 · report
fixture: returned NoiseSchedule_Scalar · level recorded 2026-09-22 f447b1a658e14845 · report
fixture: returned VDMConfig · level recorded 2026-09-22 a08f56d5383b4deb · report
fixture: returned VDMOutput · level recorded 2026-09-22 6f2b76f02949e8a2 · report
fixture: signature perm:list/1/int · returned tuple · level recorded 2026-09-22 efd5fbe71f5d4417 · report
fixture: signature value:float/0/float, dtype:str/0/str · returned function · level recorded 2026-09-22 3650a2dfe2e97298 · report
fixture: signature query:ArrayImpl/4/float, key:ArrayImpl/4/float, value:ArrayImpl/4/float, dtype:callable/0/?, bias:NoneType/0/?, axis:tuple/1/int, precision:NoneType/0/? · returned ArrayImpl · level recorded 2026-09-22 census: 11 implementations from 15 papers, 5 distinct outputs across 2 buckets, one shared input per bucket model_vdm.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant Apache-2.0 (permissive) ·
0a5352fc3e23401b · report
fixture: signature timesteps:ArrayImpl/1/float, embedding_dim:int/0/int, dtype:callable/0/? · returned ArrayImpl · level recorded 2026-09-22 32c40355362a0b32 · report
d212b472730598ed · report
fixture: returned FixedLinearSchedule · level recorded 2026-09-22 ea5520290d73f53d · report
fixture: returned LearnedLinearSchedule · level recorded 2026-09-22 abd7e936a4a2330d · report
fixture: signature mean_squared:Tensor/2/float, var:Tensor/2/float · returned Tensor · level recorded 2026-09-22 656e283a7a532e0d · report
fixture: signature batch:Tensor/2/float · returned tuple · level recorded 2026-09-22 9081710b06dc58b5 · report
fixture: signature x:Tensor/2/float, num_dims:int/0/int · returned Tensor · level recorded 2026-09-22 bb89d1300208329a · report
fixture: signature x:Tensor/1/float · returned Tensor · level recorded 2026-09-22 5c967875cb5a67f1 · report
fixture: signature x:Tensor/2/float, other:Tensor/4/float · returned Tensor · level recorded 2026-09-22 59e0526c50f31501 · report
fixture: returned Config · level recorded 2026-09-22 212e87de38149204 · report
fixture: returned LogSNR · level recorded 2026-09-22 cb2f69aa1655d268 · report
fixture: returned PosDense · level recorded 2026-09-22 3421a6b0be94af74 · report
fixture: returned WNDense · level recorded 2026-09-22 a4d6fbb9fe89feb5 · report
fixture: returned WNDilatedConv · level recorded 2026-09-22 f55545fd60b5575f · report
fixture: returned WaveNetBlock · level recorded 2026-09-22 4f4db12a92cab549 · report
fixture: signature value:float/0/float · returned function · level recorded 2026-09-22 ae2c486b1f90a559 · report
fixture: returned DiffWave · level recorded 2026-09-22 7fcce4b4613629ad · report
fixture: returned ResidualBlock · level recorded 2026-09-22 16c2edc6a8669837 · report
fixture: returned GaussianDiffusion2 · level recorded 2026-09-22 census: 14 implementations from 24 papers, 2 distinct outputs on one shared input diffusion_tf/diffusion_utils_2.py ran · our draft was wrong no licence file found · pointer only ·
5f5e07eb94d1c5d2 · report
fixture: signature mean1:SymbolicTensor/2/float, logvar1:SymbolicTensor/2/float, mean2:SymbolicTensor/2/float, logvar2:SymbolicTensor/2/float · returned SymbolicTensor · level recorded 2026-09-22 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. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.
6 assemblies on this paper: 6 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).
6 sibling assemblies for Variational Diffusion Models (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.
harvested Assembly for Variational Diffusion Models (graph method: Variational Diffusion Models). 9 bricks; individually on a fixture: 7 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.
one of 6 siblings
- VLBDiffWave entry · level 0 · unverified ·
9f2532a370c4d0c4 - Config helper · level 1 · ran ·
59e0526c50f31501 - DiffWave model · level 0 · unverified ·
dacd58542a91f9cc - LogSNR scheduler · level 1 · ran ·
212e87de38149204 - WaveNetBlock model · level 1 · ran ·
f55545fd60b5575f - PosDense helper · level 1 · ran ·
cb2f69aa1655d268 - constant helper · level 1 · ran · our draft was wrong ·
4f4db12a92cab549 - WNDense helper · level 1 · ran ·
3421a6b0be94af74 - WNDilatedConv helper · level 1 · ran ·
a4d6fbb9fe89feb5
harvested Assembly for Variational Diffusion Models (graph method: Variational Diffusion Models). 6 bricks; individually on a fixture: 6 ran (of which 5 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 6 siblings
- DiffWave entry · level 1 · ran ·
ae2c486b1f90a559 - Conv1d helper · level 1 · ran · our draft was wrong ·
2109126c17b0f297 - DiffusionEmbedding model · level 1 · ran ·
4774d75a47fa7f01 - ResidualBlock model · level 1 · ran ·
7fcce4b4613629ad - SpectrogramUpsampler model · level 1 · ran ·
e3b3ff81ba235015 - silu helper · level 1 · ran · honoured contract ·
d9caeeced554450f
harvested Assembly for Variational Diffusion Models (graph method: Variational Diffusion Models). 16 bricks; individually on a fixture: 11 ran (of which 7 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.
one of 6 siblings
- VDM entry · level 0 · unverified ·
29cb7af455c82dc4 - EncDec model · level 1 · ran ·
491caa588adc4c8f - NoiseSchedule_FixedLinear scheduler · level 1 · ran ·
069d8acfc9e74bee - NoiseSchedule_NNet scheduler · level 0 · unverified ·
6f9deb29cfceeb7f - NoiseSchedule_Scalar scheduler · level 1 · ran ·
529196025d205b76 - ScoreUNet model · level 0 · unverified ·
3c07ccd2fae8c079 - VDMConfig helper · level 1 · ran ·
f447b1a658e14845 - VDMOutput helper · level 1 · ran ·
a08f56d5383b4deb - DenseMonotone helper · level 0 · unverified ·
2bfdfec42cfdcdcf - constant_init helper · level 1 · ran · our draft was wrong ·
efd5fbe71f5d4417 - AttnBlock model · level 1 · ran ·
561db3edf2a73d49 - Base2FourierFeatures helper · level 1 · ran ·
1fd721cbc30c55b2 - ResnetBlock model · level 0 · unverified ·
32c40355362a0b32 - get_timestep_embedding helper · level 1 · ran; the contract check could not run (RAISES) ·
0a5352fc3e23401b - dot_product_attention helper · level 1 · ran; the contract check could not run (RAISES) ·
3650a2dfe2e97298 - _invert_perm helper · level 1 · ran · our draft was wrong ·
6f2b76f02949e8a2
harvested Assembly for Variational Diffusion Models (graph method: Variational Diffusion Models). 6 bricks; individually on a fixture: 5 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 6 siblings
- VDM entry · level 0 · unverified ·
d69b5df2d07b253c - FixedLinearSchedule scheduler · level 1 · ran ·
d212b472730598ed - LearnedLinearSchedule scheduler · level 1 · ran ·
ea5520290d73f53d - kl_std_normal loss · level 1 · ran · our draft was wrong ·
abd7e936a4a2330d - maybe_unpack_batch helper · level 1 · ran · our draft was wrong ·
656e283a7a532e0d - unsqueeze_right helper · level 1 · ran; the contract check could not run (RAISES) ·
9081710b06dc58b5
harvested Assembly for Variational Diffusion Models (graph method: Variational Diffusion Models). 2 bricks; individually on a fixture: 2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.
one of 6 siblings
- GaussianDiffusion2 entry · level 1 · ran ·
16c2edc6a8669837 - normal_kl loss · level 1 · ran · our draft was wrong ·
5f5e07eb94d1c5d2
harvested Assembly for Variational Diffusion Models (graph method: Variational Diffusion Models). 3 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 1 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.
one of 6 siblings
- VDM entry · level 0 · unverified ·
b780d07f50dbb622 - Discretization helper · level 1 · ran ·
bb89d1300208329a - broadcast_right helper · level 1 · ran · violated contract ·
5c967875cb5a67f1
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="2107.00630") returns these assemblies with each brick's role and wiring (how to connect).
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Density Estimation | CIFAR-10 | VDM | NLL (bits/dim) | 2.65 | #4 of 15 | Archive leaderboard | report |
| Density Estimation | ImageNet 32x32 | VDM | NLL (bits/dim) | 3.72 | #3 of 5 | Archive leaderboard | report |
| Image Generation | ImageNet 32x32 | VDM | bpd | 3.72 | #17 of 35 | Archive leaderboard | report |
| Image Generation | ImageNet 64x64 | VDM | Bits per dim | 3.40 | #40 of 65 | 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.
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