{"url":"/sota/image-generation-on-lsun-churches-256-x-256","task":{"name":"Image Generation","url":"/task/image-generation","note":null},"dataset":{"name":"LSUN Churches 256 x 256","url":"/dataset/lsun"},"category":"Computer Vision","categories":["Computer Vision","Medical","Miscellaneous","Natural Language Processing"],"category_note":null,"description":"**Image Generation** (synthesis) is the task of generating new images from an existing dataset.\r\n\r\n- **Unconditional generation** refers to generating samples unconditionally from the dataset, i.e. $p(y)$\r\n- **[Conditional image generation](/task/conditional-image-generation)** (subtask) refers to generating samples conditionally from the dataset, based on a label, i.e. $p(y|x)$.\r\n\r\nIn this section, you can find state-of-the-art leaderboards for **unconditional generation**. For conditional  generation, and other types of image generations, refer to the subtasks.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [StyleGAN](https://github.com/NVlabs/stylegan) )</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["FID","Clean-FID (trainfull)","NFE","Recall","clean-FID"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"FID":"lower","Clean-FID (trainfull)":"lower","NFE":null,"Recall":"higher","clean-FID":"lower"}},"counts":{"rows":27,"rows_with_code":26,"rows_with_paper_page":26,"rows_dated":26,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"Projected GAN","metrics":{"FID":"1.59"},"uses_additional_data":false,"paper_date":"2021-11-01","paper":"/paper/projected-gans-converge-faster","paper_url":"https://arxiv.org/abs/2111.01007v1","paper_title":"Projected GANs Converge Faster","code":"https://github.com/autonomousvision/projected_gan","n_code_links":3,"syntology":{"n_ran":38,"n_unverified":11,"n_samples":49,"n_pointer_only_licence":6}},{"rank_in_archive_order":2,"model":"Vision-aided GAN","metrics":{"Clean-FID (trainfull)":"1.72 ± 0.01","FID":"1.72"},"uses_additional_data":false,"paper_date":"2021-12-16","paper":"/paper/ensembling-off-the-shelf-models-for-gan","paper_url":"https://arxiv.org/abs/2112.09130v3","paper_title":"Ensembling Off-the-shelf Models for GAN Training","code":"https://github.com/nupurkmr9/vision-aided-gan","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":3,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"Diffusion ProjectedGAN","metrics":{"FID":"1.85"},"uses_additional_data":false,"paper_date":"2022-06-05","paper":"/paper/diffusion-gan-training-gans-with-diffusion","paper_url":"https://arxiv.org/abs/2206.02262v4","paper_title":"Diffusion-GAN: Training GANs with Diffusion","code":"https://github.com/Zhendong-Wang/Diffusion-GAN","n_code_links":4,"syntology":{"n_ran":14,"n_unverified":4,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"Patch Diffusion","metrics":{"FID":"2.66"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"CIPS","metrics":{"FID":"2.92"},"uses_additional_data":false,"paper_date":"2020-11-27","paper":"/paper/image-generators-with-conditionally","paper_url":"https://arxiv.org/abs/2011.13775v1","paper_title":"Image Generators with Conditionally-Independent Pixel Synthesis","code":"https://github.com/saic-mdal/CIPS","n_code_links":2,"syntology":null},{"rank_in_archive_order":6,"model":"StyleSwin","metrics":{"FID":"2.95"},"uses_additional_data":false,"paper_date":"2021-12-20","paper":"/paper/styleswin-transformer-based-gan-for-high-1","paper_url":"https://arxiv.org/abs/2112.10762v2","paper_title":"StyleSwin: Transformer-based GAN for High-resolution Image Generation","code":"https://github.com/microsoft/StyleSwin","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":0,"n_samples":5,"n_pointer_only_licence":3}},{"rank_in_archive_order":7,"model":"Diffusion StyleGAN2","metrics":{"FID":"3.17"},"uses_additional_data":false,"paper_date":"2022-06-05","paper":"/paper/diffusion-gan-training-gans-with-diffusion","paper_url":"https://arxiv.org/abs/2206.02262v4","paper_title":"Diffusion-GAN: Training GANs with Diffusion","code":"https://github.com/Zhendong-Wang/Diffusion-GAN","n_code_links":4,"syntology":{"n_ran":14,"n_unverified":4,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"StyleNAT","metrics":{"FID":"3.4"},"uses_additional_data":false,"paper_date":"2022-11-10","paper":"/paper/stylenat-giving-each-head-a-new-perspective","paper_url":"https://arxiv.org/abs/2211.05770v2","paper_title":"StyleNAT: Giving Each Head a New Perspective","code":"https://github.com/SHI-Labs/Neighborhood-Attention-Transformer","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"CLR-GAN","metrics":{"FID":"3.43","Recall":"0.48"},"uses_additional_data":false,"paper_date":"2024-09-30","paper":"/paper/clr-gan-improving-gans-stability-and-quality","paper_url":"https://link.springer.com/chapter/10.1007/978-3-031-73232-4_12","paper_title":"CLR-GAN: Improving GANs Stability and Quality via Consistent Latent Representation and Reconstruction","code":"https://github.com/Petecheco/CLR-GAN","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"StyleGAN2","metrics":{"Clean-FID (trainfull)":"4.28 ± 0.03","FID":"3.86"},"uses_additional_data":false,"paper_date":"2019-12-03","paper":"/paper/analyzing-and-improving-the-image-quality-of","paper_url":"https://arxiv.org/abs/1912.04958v2","paper_title":"Analyzing and Improving the Image Quality of StyleGAN","code":"https://github.com/labmlai/annotated_deep_learning_paper_implementations","n_code_links":126,"syntology":{"n_ran":60,"n_unverified":32,"n_samples":92,"n_pointer_only_licence":18}},{"rank_in_archive_order":11,"model":"Polarity-StyleGAN2","metrics":{"FID":"3.92"},"uses_additional_data":false,"paper_date":"2022-03-03","paper":"/paper/polarity-sampling-quality-and-diversity","paper_url":"https://arxiv.org/abs/2203.01993v2","paper_title":"Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values","code":"https://github.com/AhmedImtiazPrio/magnet-polarity","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"TDPM+ (TTrunc=99)","metrics":{"FID":"3.98","NFE":"100"},"uses_additional_data":false,"paper_date":"2022-02-19","paper":"/paper/truncated-diffusion-probabilistic-models","paper_url":"https://arxiv.org/abs/2202.09671v4","paper_title":"Truncated Diffusion Probabilistic Models and Diffusion-based Adversarial Auto-Encoders","code":"https://github.com/jegzheng/truncated-diffusion-probabilistic-models","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"INR-GAN-bil","metrics":{"FID":"4.04"},"uses_additional_data":false,"paper_date":"2020-11-24","paper":"/paper/adversarial-generation-of-continuous-images","paper_url":"https://arxiv.org/abs/2011.12026v2","paper_title":"Adversarial Generation of Continuous Images","code":"https://github.com/universome/inr-gan","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"Unleashing Transformers","metrics":{"FID":"4.07"},"uses_additional_data":false,"paper_date":"2021-11-24","paper":"/paper/unleashing-transformers-parallel-token","paper_url":"https://arxiv.org/abs/2111.12701v1","paper_title":"Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes","code":"https://github.com/samb-t/unleashing-transformers","n_code_links":3,"syntology":null},{"rank_in_archive_order":15,"model":"StyleGAN","metrics":{"Clean-FID (trainfull)":"4.75 ± 0.01","FID":"4.21"},"uses_additional_data":false,"paper_date":"2018-12-12","paper":"/paper/a-style-based-generator-architecture-for","paper_url":"http://arxiv.org/abs/1812.04948v3","paper_title":"A Style-Based Generator Architecture for Generative Adversarial Networks","code":"https://github.com/NVlabs/stylegan","n_code_links":83,"syntology":{"n_ran":229,"n_unverified":167,"n_samples":396,"n_pointer_only_licence":229}},{"rank_in_archive_order":16,"model":"SWAGAN-Bi","metrics":{"FID":"4.97"},"uses_additional_data":false,"paper_date":"2021-02-11","paper":"/paper/swagan-a-style-based-wavelet-driven","paper_url":"https://arxiv.org/abs/2102.06108v1","paper_title":"SWAGAN: A Style-based Wavelet-driven Generative Model","code":"https://github.com/rinongal/swagan","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":17,"model":"WaveDiff","metrics":{"FID":"5.06","NFE":"4","Recall":"0.40"},"uses_additional_data":false,"paper_date":"2022-11-29","paper":"/paper/wavelet-diffusion-models-are-fast-and","paper_url":"https://arxiv.org/abs/2211.16152v2","paper_title":"Wavelet Diffusion Models are fast and scalable Image Generators","code":"https://github.com/vinairesearch/wavediff","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"MSG-StyleGAN","metrics":{"Clean-FID (trainfull)":"5.38 ± 0.03","FID":"5.2"},"uses_additional_data":false,"paper_date":"2019-03-14","paper":"/paper/msg-gan-multi-scale-gradients-gan-for-more","paper_url":"https://arxiv.org/abs/1903.06048v4","paper_title":"MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks","code":"https://github.com/akanimax/BMSG-GAN","n_code_links":5,"syntology":{"n_ran":14,"n_unverified":1,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"DDGAN","metrics":{"FID":"5.25"},"uses_additional_data":false,"paper_date":"2021-12-15","paper":"/paper/tackling-the-generative-learning-trilemma-1","paper_url":"https://arxiv.org/abs/2112.07804v2","paper_title":"Tackling the Generative Learning Trilemma with Denoising Diffusion GANs","code":"https://github.com/NVlabs/denoising-diffusion-gan","n_code_links":5,"syntology":{"n_ran":6,"n_unverified":6,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":20,"model":"LFM","metrics":{"FID":"5.54"},"uses_additional_data":false,"paper_date":"2023-07-17","paper":"/paper/flow-matching-in-latent-space","paper_url":"https://arxiv.org/abs/2307.08698v1","paper_title":"Flow Matching in Latent Space","code":"https://github.com/vinairesearch/lfm","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":21,"model":"PGGAN","metrics":{"Clean-FID (trainfull)":"6.43 ± 0.05","FID":"6.42"},"uses_additional_data":false,"paper_date":"2017-10-27","paper":"/paper/progressive-growing-of-gans-for-improved","paper_url":"http://arxiv.org/abs/1710.10196v3","paper_title":"Progressive Growing of GANs for Improved Quality, Stability, and Variation","code":"https://github.com/tkarras/progressive_growing_of_gans","n_code_links":115,"syntology":{"n_ran":26,"n_unverified":63,"n_samples":89,"n_pointer_only_licence":19}},{"rank_in_archive_order":22,"model":"Denoising Diffusion Probabilistic Model","metrics":{"FID":"7.89"},"uses_additional_data":false,"paper_date":"2020-06-19","paper":"/paper/denoising-diffusion-probabilistic-models","paper_url":"https://arxiv.org/abs/2006.11239v2","paper_title":"Denoising Diffusion Probabilistic Models","code":"https://github.com/labmlai/annotated_deep_learning_paper_implementations","n_code_links":70,"syntology":{"n_ran":186,"n_unverified":67,"n_samples":253,"n_pointer_only_licence":62}},{"rank_in_archive_order":23,"model":"PNDM","metrics":{"FID":"8.69"},"uses_additional_data":false,"paper_date":"2022-02-20","paper":"/paper/pseudo-numerical-methods-for-diffusion-models-1","paper_url":"https://arxiv.org/abs/2202.09778v2","paper_title":"Pseudo Numerical Methods for Diffusion Models on Manifolds","code":"https://github.com/sanster/lama-cleaner","n_code_links":9,"syntology":{"n_ran":10,"n_unverified":4,"n_samples":14,"n_pointer_only_licence":2}},{"rank_in_archive_order":24,"model":"TransGAN","metrics":{"FID":"8.94"},"uses_additional_data":false,"paper_date":"2021-02-14","paper":"/paper/transgan-two-transformers-can-make-one-strong","paper_url":"https://arxiv.org/abs/2102.07074v4","paper_title":"TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up","code":"https://github.com/VITA-Group/TransGAN","n_code_links":10,"syntology":{"n_ran":8,"n_unverified":0,"n_samples":8,"n_pointer_only_licence":7}},{"rank_in_archive_order":25,"model":"VE (erel=0.01)","metrics":{"FID":"26.46"},"uses_additional_data":false,"paper_date":"2021-05-28","paper":"/paper/gotta-go-fast-when-generating-data-with-score","paper_url":"https://arxiv.org/abs/2105.14080v1","paper_title":"Gotta Go Fast When Generating Data with Score-Based Models","code":"https://github.com/AlexiaJM/score_sde_fast_sampling","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":26,"model":"VE (erel=0.02)","metrics":{"FID":"26.46"},"uses_additional_data":false,"paper_date":"2021-05-28","paper":"/paper/gotta-go-fast-when-generating-data-with-score","paper_url":"https://arxiv.org/abs/2105.14080v1","paper_title":"Gotta Go Fast When Generating Data with Score-Based Models","code":"https://github.com/AlexiaJM/score_sde_fast_sampling","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":27,"model":"BOSS","metrics":{"clean-FID":"13.21"},"uses_additional_data":false,"paper_date":"2023-12-27","paper":"/paper/bellman-optimal-step-size-straightening-of","paper_url":"https://arxiv.org/abs/2312.16414v3","paper_title":"Bellman Optimal Stepsize Straightening of Flow-Matching Models","code":"https://github.com/nguyenngocbaocmt02/boss","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":6,"n_samples":14,"n_pointer_only_licence":14}}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":20,"rows_with_any_sample_ran":20,"distinct_papers_with_graph_line":18,"distinct_papers_with_any_sample_ran":18,"samples_over_distinct_papers":{"n_ran":626,"n_unverified":365,"n_samples":991,"n_pointer_only_licence":367,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":642,"n_unverified":370,"n_samples":1012,"n_pointer_only_licence":370,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}