Papers › SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis

SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis

4 Jul 2023arXiv:2307.01952archive 2025-07-28

Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas Müller, Joe Penna, Robin Rombach

We present SDXL, a latent diffusion model for text-to-image synthesis. Compared to previous versions of Stable Diffusion, SDXL leverages a three times larger UNet backbone: The increase of model parameters is mainly due to more attention blocks and a larger cross-attention context as SDXL uses a second text encoder. We design multiple novel conditioning schemes and train SDXL on multiple aspect ratios. We also introduce a refinement model which is used to improve the visual fidelity of samples generated by SDXL using a post-hoc image-to-image technique. We demonstrate that SDXL shows drastically improved performance compared the previous versions of Stable Diffusion and achieves results competitive with those of black-box state-of-the-art image generators. In the spirit of promoting open research and fostering transparency in large model training and evaluation, we provide access to code and model weights at https://github.com/Stability-AI/generative-models

PaperPDFCodeCode 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 27 code samples harvested from 3 repositories linked to this paper: 14 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 0 where Syntology's instrument failed) · 13 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 14 samples from 1 repository, 4 ran (4 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 13 samples from 2 repositories, 10 ran (10 with no instrument failure; 0 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 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

stability-ai/generative-models officialmentioned in papermentioned on GitHubpytorchSyntology: 14 samples, 4 ran (0 instrument-failed) · 10 unverified report
andrew-miao/RPO mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Subject-driven Text-to-Image Generation via Preference-based Reinforcement Learning (arXiv:2407.12164; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
benearnthof/fm_boosting mentioned on GitHubpytorchSyntology: not harvested report
bytedance/cascadev mentioned on GitHubpytorchAGPL-3.0Syntology: not harvested report
compvis/fm-boosting mentioned on GitHubpytorchSyntology: 9 samples, 7 ran (0 instrument-failed) · 2 unverified report
tillmannohm/fruit-SALAD mentioned on GitHubpytorchSyntology: not harvested report
vision-xl/codes mentioned on GitHubpytorchSyntology: not harvested report
wellzline/protip mentioned on GitHubpytorchSyntology: not harvested report
yuchen413/text2image_safety mentioned on GitHubpytorchMITSyntology: 4 samples, 3 ran (0 instrument-failed) · 1 unverified 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

27 samples harvested: 14 ran (of which 0 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 0 where Syntology's instrument failed) · 13 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.

14ran
13unverified

Licence: 0 of the 27 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”. “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.

stability-ai/generative-models official repository 14 samples, 4 ran (0 instrument-failed) · 10 unverified
AlphaBlender sgm/modules/diffusionmodules/openaimodel.py ran MIT (permissive) · ad950c2482098461 · report fixture: signature x_spatial:Tensor/3/float, x_temporal:Tensor/3/float, image_only_indicator:Tensor/2/float · returned Tensor · level recorded 2026-09-24
Downsample in 58 papers sgm/modules/diffusionmodules/openaimodel.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 3ee1e71fc45df15b · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
Timestep sgm/modules/diffusionmodules/openaimodel.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 9eb2201b9e41a2d3 · report fixture: signature t:Tensor/1/float · returned Tensor · level recorded 2026-09-22
Upsample in 77 papers sgm/modules/diffusionmodules/openaimodel.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 184b9747aa868755 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
BasicTransformerBlock in 22 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 40d8ba71b73d29d4 · report
CrossAttention in 85 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 75e42984360b91c5 · report
MemoryEfficientCrossAttention in 7 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 5cbc6286835eafa2 · report
SpatialTransformer in 23 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 9778f92ec0ba682d · report
SpatialVideoTransformer sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · d0977bbe6d3f1b49 · report
TimestepBlock in 11 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 579d378f598c87ed · report
TimestepEmbedSequential in 28 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 0b73a46bc27e47c6 · report
UNetModel in 23 papers sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 5c5d69427c0df4ac · report
VideoTransformerBlock sgm/modules/diffusionmodules/openaimodel.py unverified MIT (permissive) · 54cb1b411b5423e1 · report
compvis/fm-boosting community (archive-listed) 9 samples, 7 ran (0 instrument-failed) · 2 unverified
Downsample in 58 papers fmboost/models/unet/model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 03cbae4dc00dbd28 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
LinearQKVAttention fmboost/models/unet/model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 5dc3f2495524ee28 · report fixture: signature q:Tensor/4/float, k:Tensor/4/float, v:Tensor/4/float · returned Tensor · level recorded 2026-09-22
Pool2d fmboost/models/unet/model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 7340098015ec809d · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
QKVAttention in 25 papers fmboost/models/unet/model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 0bb33f06a6461c7a · report fixture: signature q:Tensor/4/float, k:Tensor/4/float, v:Tensor/4/float · returned Tensor · level recorded 2026-09-22
ResBlock in 127 papers
census: 6 implementations from 5 papers, 6 distinct outputs across 2 buckets, one shared input per bucket
fmboost/models/unet/model.py ran MIT (permissive) · 48e8939cab1b71e2 · report fixture: signature x:Tensor/4/float, emb:Tensor/2/float · returned Tensor · level recorded 2026-09-24
UnPool2d fmboost/models/unet/model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 8e7f07ae79ee1de0 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
Upsample in 77 papers fmboost/models/unet/model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 63b3160a59e80ddb · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
EfficientUNet fmboost/models/unet/model.py unverified MIT (permissive) · 173b3ff1d8ad06a9 · report
SpatialSelfAttention fmboost/models/unet/model.py unverified MIT (permissive) · ffc4ff1275bb5b7d · report
yuchen413/text2image_safety community (archive-listed) 4 samples, 3 ran (0 instrument-failed) · 1 unverified
get_dictionary search_utils.py ran MIT (permissive) · 8b2e0a95b5d28430 · report fixture: signature len_subword:int/0/int, en:bool/0/bool · returned tuple · level recorded 2026-09-26
load_data in 700 papers search_utils.py ran MIT (permissive) · 73a61f7a5d72f3e7 · report fixture: signature path:str/0/str · returned list · level recorded 2026-09-26
load_nsfw_prompts baseline/baseline.py ran MIT (permissive) · b85c3309a7d80fdd · report fixture: signature filepath:str/0/str · returned tuple · level recorded 2026-09-26

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.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for SDXL (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for SDXL (graph method: SDXL). 13 bricks; individually on a fixture: 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. EfficientUNet entry · level 0 · unverified · 173b3ff1d8ad06a9
  2. Downsample helper · level 1 · ran · 03cbae4dc00dbd28
  3. Pool2d helper · level 1 · ran · 7340098015ec809d
  4. ResBlock helper · level 1 · ran · 48e8939cab1b71e2
  5. SpatialSelfAttention model · level 0 · unverified · ffc4ff1275bb5b7d
  6. TimestepEmbedSequential helper · level 0 · unverified · d9ce54d68567e11c
  7. UnPool2d helper · level 1 · ran · 8e7f07ae79ee1de0
  8. Upsample helper · level 1 · ran · 63b3160a59e80ddb
  9. timestep_embedding helper · level 1 · ran · our draft was wrong · 270cd726f1952c2a
  10. zero_module helper · level 1 · ran · our draft was wrong · 129b804760b3115f
  11. TimestepBlock helper · level 0 · unverified · 7958a01675377ce4
  12. LinearQKVAttention model · level 1 · ran · 5dc3f2495524ee28
  13. QKVAttention model · level 1 · ran · 0bb33f06a6461c7a

harvested Assembly for SDXL (graph method: SDXL). 28 bricks; individually on a fixture: 15 ran (of which 4 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 2 violated, 8 with no contract checked; 5 where Syntology's instrument failed) · 13 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. UNetModel entry · level 0 · unverified · 5c5d69427c0df4ac
  2. Downsample model · level 1 · ran · 3ee1e71fc45df15b
  3. ResBlock model · level 0 · unverified · 77ab8d10090cbbae
  4. SpatialTransformer helper · level 0 · unverified · 9778f92ec0ba682d
  5. Timestep helper · level 1 · ran · 9eb2201b9e41a2d3
  6. TimestepEmbedSequential model · level 0 · unverified · 0b73a46bc27e47c6
  7. Upsample model · level 1 · ran · 184b9747aa868755
  8. conv_nd helper · level 0 · unverified · fe4eb545bbb728e0
  9. exists helper · level 1 · ran · violated contract · aa5486a3650902d8
  10. linear helper · level 1 · ran · our draft was wrong · 0e30585f04efffbe
  11. normalization helper · level 1 · ran · our draft was wrong · daafc1c89511b8be
  12. timestep_embedding helper · level 1 · ran; the contract check could not run · 1b261a79cdace383
  13. zero_module helper · level 1 · ran · our draft was wrong · 129b804760b3115f
  14. avg_pool_nd helper · level 0 · unverified · ecd0fc28815b65ae
  15. TimestepBlock helper · level 0 · unverified · 579d378f598c87ed
  16. checkpoint helper · level 0 · unverified · c596e50230fd9ed8
  17. BasicTransformerBlock helper · level 0 · unverified · 40d8ba71b73d29d4
  18. Normalize helper · level 1 · ran · our draft was wrong · c3a6b977022957cb
  19. SpatialVideoTransformer helper · level 0 · unverified · d0977bbe6d3f1b49
  20. GroupNorm32 helper · level 1 · ran · 426c2385fa431d9a
  21. CheckpointFunction helper · level 1 · ran · 70ec8e3dd3981b5e
  22. CrossAttention helper · level 0 · unverified · 75e42984360b91c5
  23. FeedForward helper · level 1 · ran · 3abfca8077fe9d76
  24. MemoryEfficientCrossAttention helper · level 0 · unverified · 5cbc6286835eafa2
  25. AlphaBlender helper · level 1 · ran · ad950c2482098461
  26. VideoTransformerBlock helper · level 0 · unverified · 54cb1b411b5423e1
  27. default helper · level 1 · ran · violated contract · 424012cb37b31172
  28. GEGLU helper · level 1 · ran · 387673fa5780bf50

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="2307.01952") 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
WISE SDXL Biology 0.44 #7 of 11 Archive leaderboard report
WISE SDXL Chemistry 0.27 #7 of 11 Archive leaderboard report
WISE SDXL Cultural 0.43 #7 of 11 Archive leaderboard report
WISE SDXL Overall 0.43 #7 of 11 Archive leaderboard report
WISE SDXL Physics 0.45 #7 of 11 Archive leaderboard report
WISE SDXL Space 0.47 #7 of 11 Archive leaderboard report
WISE SDXL Time 0.48 #7 of 11 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Biology 0.44 #9 of 14 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Chemistry 0.27 #9 of 14 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Cultural 0.43 #9 of 14 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Overall 0.43 #9 of 14 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Physics 0.45 #9 of 14 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Space 0.47 #9 of 14 Archive leaderboard report
Image Generation WISE stable-diffusion-xl-base-0.9 Time 0.48 #9 of 14 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

DiffusionLatent Diffusion Model

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