Papers › Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion

Kandinsky: an Improved Text-to-Image Synthesis with Image Prior and Latent Diffusion

5 Oct 2023arXiv:2310.03502archive 2025-07-28

Anton Razzhigaev, Arseniy Shakhmatov, Anastasia Maltseva, Vladimir Arkhipkin, Igor Pavlov, Ilya Ryabov, Angelina Kuts, Alexander Panchenko, Andrey Kuznetsov, Denis Dimitrov

Text-to-image generation is a significant domain in modern computer vision and has achieved substantial improvements through the evolution of generative architectures. Among these, there are diffusion-based models that have demonstrated essential quality enhancements. These models are generally split into two categories: pixel-level and latent-level approaches. We present Kandinsky1, a novel exploration of latent diffusion architecture, combining the principles of the image prior models with latent diffusion techniques. The image prior model is trained separately to map text embeddings to image embeddings of CLIP. Another distinct feature of the proposed model is the modified MoVQ implementation, which serves as the image autoencoder component. Overall, the designed model contains 3.3B parameters. We also deployed a user-friendly demo system that supports diverse generative modes such as text-to-image generation, image fusion, text and image fusion, image variations generation, and text-guided inpainting/outpainting. Additionally, we released the source code and checkpoints for the Kandinsky models. Experimental evaluations demonstrate a FID score of 8.03 on the COCO-30K dataset, marking our model as the top open-source performer in terms of measurable image generation quality.

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 18 code samples harvested from 3 repositories linked to this paper: 12 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 0 violated, 2 with no contract checked; 7 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 4 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: 8 samples from 1 repository, 3 ran (2 with no instrument failure; 1 where Syntology's instrument failed); found in paper text by Syntology: 10 samples from 2 repositories, 9 ran (3 with no instrument failure; 6 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 12 samples that ran has no recorded date.

ai-forever/Kandinsky-2 officialmentioned in paperpytorchSyntology: 8 samples, 3 ran (1 instrument-failed) · 5 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

18 samples harvested: 12 ran (of which 1 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 0 violated, 2 with no contract checked; 7 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
4ran · our draft was wrong
3ran; the contract check could not run
2ran
6unverified

Licence: 10 of the 18 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 1 sample 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.

ai-forever/Kandinsky-2 official repository 8 samples, 3 ran (1 instrument-failed) · 5 unverified
AttentionPooling in 9 papers kandinsky2/model/text2im_model2_1.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · f60bca05893fa428 · report fixture: returned AttentionPooling · level recorded 2026-09-22
QKVAttention in 25 papers kandinsky2/model/text2im_model2_1.py ran fingerprinted Apache-2.0 (permissive) · 317687cb15c56f85 · report fixture: signature qkv:Tensor/3/float, encoder_kv:NoneType/0/? · returned Tensor · level recorded 2026-09-24
attention in 144 papers
census: 11 implementations from 10 papers, 7 distinct outputs across 5 buckets, one shared input per bucket this sample: bucket 3 of 5, arg 1: rank 3, kind float, dtype float32 · arg 2: rank 3, kind float, dtype float32 · arg 3: rank 3, kind float, dtype float32
kandinsky2/model/text2im_model2_1.py ran · our draft was wrong Apache-2.0 (permissive) · fe30479834a53868 · report fixture: signature q:Tensor/3/float, k:Tensor/3/float, v:Tensor/3/float, d_k:int/0/int · returned Tensor · level recorded 2026-09-22
AttentionBlock in 69 papers kandinsky2/model/text2im_model2_1.py unverified Apache-2.0 (permissive) · 88a5c5be88a2747b · report
Text2ImUNet kandinsky2/model/text2im_model2_1.py unverified Apache-2.0 (permissive) · 7759535afd6da808 · report
TimestepEmbedSequential in 28 papers kandinsky2/model/text2im_model2_1.py unverified Apache-2.0 (permissive) · e587c8596212bcdf · report
UNetModel in 23 papers kandinsky2/model/text2im_model2_1.py unverified Apache-2.0 (permissive) · 820bf30db90e06e6 · report
ai-forever/MoVQGAN found in paper text by Syntology 1 sample, 0 ran · 1 unverified
MOVQ movqgan/models/vqgan.py unverified no licence file found · pointer only · 47ec7613fe517e5e · report
deep-floyd/IF found in paper text by Syntology 9 samples, 9 ran (6 instrument-failed) · 0 unverified
approx_standard_normal_cdf in 150 papers
census: 4 implementations from 14 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
deepfloyd_if/model/losses.py ran · honoured contract fingerprinted no licence file found · pointer only · 9deab2a47adbf3b7 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
betas_for_alpha_bar in 123 papers deepfloyd_if/model/gaussian_diffusion.py ran · honoured contract no licence file found · pointer only · 6cdc8772f5db6f48 · report fixture: signature num_diffusion_timesteps:int/0/int, alpha_bar:callable/0/?, max_beta:float/0/float · returned ndarray · level recorded 2026-09-22
discretized_gaussian_log_likelihood in 124 papers
census: 6 implementations from 10 papers, 2 distinct outputs on one shared input
deepfloyd_if/model/losses.py ran · our draft was wrong no licence file found · pointer only · f702cecaa7981c73 · report fixture: signature x:Tensor/2/float, means:Tensor/2/float, log_scales:Tensor/2/float · returned Tensor · level recorded 2026-09-24
gelu in 308 papers
census: 43 implementations from 74 papers, 6 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
deepfloyd_if/model/nn.py ran · our draft was wrong fingerprinted no licence file found · pointer only · f51cf2cbd3e598d8 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-22
gelu_jit deepfloyd_if/model/nn.py ran; the contract check could not run (RAISES) no licence file found · pointer only · cd23a508b079e2d4 · report fixture: returned Tensor · level recorded 2026-09-22
get_named_beta_schedule in 119 papers deepfloyd_if/model/gaussian_diffusion.py ran; the contract check could not run (RAISES) · metamorphic tier: well formed no licence file found · pointer only · 141937afc98f8d1e · report fixture: signature schedule_name:str/0/str, num_diffusion_timesteps:int/0/int · returned ndarray · level recorded 2026-09-22
mean_flat in 53 papers
census: 9 implementations from 18 papers, 4 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 3, kind float, dtype float32
deepfloyd_if/model/nn.py ran · our draft was wrong fingerprinted no licence file found · pointer only · f6d7c009a8efb8b7 · report fixture: signature tensor:Tensor/3/float · returned Tensor · level recorded, date not recorded
normal_kl in 188 papers
census: 14 implementations from 24 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32 · arg 3: rank 2, kind float, dtype float32 · arg 4: rank 2, kind float, dtype float32
deepfloyd_if/model/losses.py ran · honoured contract fingerprinted no licence file found · pointer only · c3bf3f13d5f63d5b · report fixture: signature mean1:Tensor/2/float, logvar1:Tensor/2/float, mean2:Tensor/2/float, logvar2:Tensor/2/float · returned Tensor · level recorded 2026-09-22
space_timesteps in 72 papers deepfloyd_if/model/respace.py ran; the contract check could not run (RAISES) no licence file found · pointer only · 4e881da877b25beb · report fixture: signature num_timesteps:int/0/int, section_counts:list/1/int · returned set · 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. 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 Kandinsky (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Kandinsky (graph method: Kandinsky). 20 bricks; individually on a fixture: 10 ran (of which 4 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 10 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. Text2ImUNet entry · level 0 · unverified · 7759535afd6da808
  2. AttentionPooling helper · level 1 · ran · f60bca05893fa428
  3. UNetModel model · level 0 · unverified · 820bf30db90e06e6
  4. timestep_embedding helper · level 1 · ran · our draft was wrong · 270cd726f1952c2a
  5. attention helper · level 1 · ran · our draft was wrong · fe30479834a53868
  6. AttentionBlock helper · level 0 · unverified · 88a5c5be88a2747b
  7. Downsample helper · level 1 · ran · 8c5676227adb2492
  8. ResBlock helper · level 0 · unverified · 99ce6c633e8c9151
  9. TimestepEmbedSequential helper · level 0 · unverified · e587c8596212bcdf
  10. Upsample helper · level 1 · ran · 6998a3677700ee6c
  11. conv_nd helper · level 0 · unverified · fe4eb545bbb728e0
  12. convert_module_to_f16 helper · level 0 · unverified · ac07425399caf784
  13. convert_module_to_f32 helper · level 0 · unverified · b283a8ab305dc007
  14. linear helper · level 1 · ran · our draft was wrong · 0e30585f04efffbe
  15. normalization helper · level 1 · ran · our draft was wrong · c93ddadd4581c6e6
  16. zero_module helper · level 1 · ran · our draft was wrong · 129b804760b3115f
  17. QKVAttention helper · level 1 · ran · 317687cb15c56f85
  18. avg_pool_nd helper · level 0 · unverified · ecd0fc28815b65ae
  19. TimestepBlock helper · level 0 · unverified · 7958a01675377ce4
  20. GroupNorm32 helper · level 1 · ran · afad4c14b97816a5

harvested Assembly for Kandinsky (graph method: Kandinsky). 12 bricks; individually on a fixture: 9 ran (of which 5 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. MOVQ entry · level 0 · unverified · 47ec7613fe517e5e
  2. Encoder helper · level 1 · ran · 63c2a9fa53a4e807
  3. LitEma helper · level 1 · ran · 5b6d9f0747d7c874
  4. MOVQDecoder helper · level 0 · unverified · dc9e571552633f72
  5. instantiate_from_config helper · level 1 · ran · our draft was wrong · 56014014c4ed0790
  6. AttnBlock model · level 0 · unverified · 5c22fcb7b2c1bd3b
  7. Downsample model · level 1 · ran · 37124eedd502424a
  8. Normalize helper · level 1 · ran · our draft was wrong · c3a6b977022957cb
  9. ResnetBlock model · level 1 · ran · 4c7c8720519325ed
  10. nonlinearity helper · level 1 · ran · our draft was wrong · 3137073275f8c21a
  11. Upsample model · level 1 · ran · 38a1b1b7d030df65
  12. get_obj_from_str helper · level 1 · ran · our draft was wrong · 4e4f3bef7e8672e8

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="2310.03502") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image GenerationText to Image GenerationText-to-Image Generation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Text-to-Image Generation COCO (Common Objects in Context) Kandinsky FID 8.03 #19 of 69 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

CLIPDiffusionFAVOR+Performer

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