{"url":"/sota/image-generation-on-imagenet-256x256","task":{"name":"Image Generation","url":"/task/image-generation","note":null},"dataset":{"name":"ImageNet 256x256","url":null},"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","Inception score","NFE"],"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","Inception score":"higher","NFE":null}},"counts":{"rows":94,"rows_with_code":85,"rows_with_paper_page":91,"rows_dated":91,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"SiT-XL/2 + UCGM-S (E2E-VAE + 40 sampling steps + CFG)","metrics":{"FID":"1.06","NFE":"80"},"uses_additional_data":false,"paper_date":"2025-05-12","paper":"/paper/unified-continuous-generative-models","paper_url":"https://arxiv.org/abs/2505.07447v2","paper_title":"Unified Continuous Generative Models","code":"https://github.com/LINs-Lab/UCGM","n_code_links":1,"syntology":{"n_ran":16,"n_unverified":5,"n_samples":21,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"UCGM-XL/2 (VA-VAE + 30 sampling steps, without guidance)","metrics":{"FID":"1.21","NFE":"30"},"uses_additional_data":false,"paper_date":"2025-05-12","paper":"/paper/unified-continuous-generative-models","paper_url":"https://arxiv.org/abs/2505.07447v2","paper_title":"Unified Continuous Generative Models","code":"https://github.com/LINs-Lab/UCGM","n_code_links":1,"syntology":{"n_ran":16,"n_unverified":5,"n_samples":21,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"UCGM-XL/2 (E2E-VAE + 40 sampling steps, without guidance)","metrics":{"FID":"1.21","NFE":"40"},"uses_additional_data":false,"paper_date":"2025-05-12","paper":"/paper/unified-continuous-generative-models","paper_url":"https://arxiv.org/abs/2505.07447v2","paper_title":"Unified Continuous Generative Models","code":"https://github.com/LINs-Lab/UCGM","n_code_links":1,"syntology":{"n_ran":16,"n_unverified":5,"n_samples":21,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"EDM2-L + DDO (SD-VAE, 25 steps, DPM-Solver-v3)","metrics":{"FID":"1.21","NFE":"50"},"uses_additional_data":false,"paper_date":"2025-03-03","paper":"/paper/direct-discriminative-optimization-your-1","paper_url":"https://arxiv.org/abs/2503.01103v2","paper_title":"Direct Discriminative Optimization: Your Likelihood-Based Visual Generative Model is Secretly a GAN Discriminator","code":"https://github.com/nvlabs/ddo","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"LightningDiT + UCGM-S (VA-VAE + 50 sampling steps + CFG)","metrics":{"FID":"1.21","NFE":"100"},"uses_additional_data":false,"paper_date":"2025-05-12","paper":"/paper/unified-continuous-generative-models","paper_url":"https://arxiv.org/abs/2505.07447v2","paper_title":"Unified Continuous Generative Models","code":"https://github.com/LINs-Lab/UCGM","n_code_links":1,"syntology":{"n_ran":16,"n_unverified":5,"n_samples":21,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"xAR-H","metrics":{"FID":"1.24"},"uses_additional_data":false,"paper_date":"2025-02-27","paper":"/paper/beyond-next-token-next-x-prediction-for","paper_url":"https://arxiv.org/abs/2502.20388v1","paper_title":"Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation","code":"https://github.com/OliverRensu/xAR","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"SiT-XL/2 + REPA-E","metrics":{"FID":"1.26","Inception score":"314.9"},"uses_additional_data":false,"paper_date":"2025-04-15","paper":"/paper/repa-e-unlocking-vae-for-end-to-end-tuning-of","paper_url":"https://arxiv.org/abs/2504.10483","paper_title":"REPA-E: Unlocking VAE for End-to-End Tuning of Latent Diffusion Transformers","code":"https://github.com/End2End-Diffusion/REPA-E","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":8,"model":"DDT-XL/2(22en6de 675M + guidance interval )","metrics":{"FID":"1.26","Inception score":"310.6"},"uses_additional_data":false,"paper_date":"2025-04-08","paper":"/paper/ddt-decoupled-diffusion-transformer-1","paper_url":"https://arxiv.org/abs/2504.05741v2","paper_title":"DDT: Decoupled Diffusion Transformer","code":"https://github.com/MCG-NJU/DDT","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"xAR-L","metrics":{"FID":"1.28"},"uses_additional_data":false,"paper_date":"2025-02-27","paper":"/paper/beyond-next-token-next-x-prediction-for","paper_url":"https://arxiv.org/abs/2502.20388v1","paper_title":"Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation","code":"https://github.com/OliverRensu/xAR","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"FACM (2-step)","metrics":{"FID":"1.32","NFE":"2"},"uses_additional_data":false,"paper_date":"2025-07-04","paper":"/paper/flow-anchored-consistency-models","paper_url":"https://arxiv.org/abs/2507.03738v1","paper_title":"Flow-Anchored Consistency Models","code":"https://github.com/ali-vilab/FACM","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":6,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"GMem  (with the guidance interval)","metrics":{"FID":"1.32"},"uses_additional_data":false,"paper_date":"2024-12-11","paper":"/paper/generative-modeling-with-explicit-memory","paper_url":"https://arxiv.org/abs/2412.08781v1","paper_title":"Generative Modeling with Explicit Memory","code":"https://github.com/lins-lab/gmem","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":7,"n_samples":15,"n_pointer_only_licence":15}},{"rank_in_archive_order":12,"model":"SiT-XL/2 + MG","metrics":{"FID":"1.34"},"uses_additional_data":false,"paper_date":"2025-02-17","paper":"/paper/diffusion-models-without-classifier-free-1","paper_url":"https://arxiv.org/abs/2502.12154v1","paper_title":"Diffusion Models without Classifier-free Guidance","code":"https://github.com/lucidrains/classifier-free-guidance-pytorch","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":4,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"AliTok-XL, autoregressive, 662M","metrics":{"FID":"1.35","Inception score":"318.8"},"uses_additional_data":false,"paper_date":"2025-06-05","paper":"/paper/alitok-towards-sequence-modeling-alignment-1","paper_url":"https://arxiv.org/abs/2506.05289v1","paper_title":"AliTok: Towards Sequence Modeling Alignment between Tokenizer and Autoregressive Model","code":"https://github.com/ali-vilab/alitok","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":4,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":14,"model":"LightningDiT + VA-VAE (with the guidance interval)","metrics":{"FID":"1.35"},"uses_additional_data":false,"paper_date":"2025-01-02","paper":"/paper/reconstruction-vs-generation-taming-1","paper_url":"https://arxiv.org/abs/2501.01423v2","paper_title":"Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models","code":"https://github.com/hustvl/LightningDiT","n_code_links":2,"syntology":null},{"rank_in_archive_order":15,"model":"SiD2","metrics":{"FID":"1.38"},"uses_additional_data":false,"paper_date":"2024-10-25","paper":"/paper/simpler-diffusion-sid2-1-5-fid-on-imagenet512","paper_url":"https://arxiv.org/abs/2410.19324v2","paper_title":"Simpler Diffusion (SiD2): 1.5 FID on ImageNet512 with pixel-space diffusion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"SiT↓-XL/2+U-REPA (with the guidance interval)","metrics":{"FID":"1.41"},"uses_additional_data":false,"paper_date":"2025-03-24","paper":"/paper/u-repa-aligning-diffusion-u-nets-to-vits","paper_url":"https://arxiv.org/abs/2503.18414v1","paper_title":"U-REPA: Aligning Diffusion U-Nets to ViTs","code":"https://github.com/yuchuantian/u-repa","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"AliTok-XL, autoregressive, 318M","metrics":{"FID":"1.42","Inception score":"326.6"},"uses_additional_data":false,"paper_date":"2025-06-05","paper":"/paper/alitok-towards-sequence-modeling-alignment-1","paper_url":"https://arxiv.org/abs/2506.05289v1","paper_title":"AliTok: Towards Sequence Modeling Alignment between Tokenizer and Autoregressive Model","code":"https://github.com/ali-vilab/alitok","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":4,"n_samples":8,"n_pointer_only_licence":8}},{"rank_in_archive_order":18,"model":"SiT-XL/2 + REPA (with the guidance interval)","metrics":{"FID":"1.42"},"uses_additional_data":false,"paper_date":"2024-10-09","paper":"/paper/representation-alignment-for-generation","paper_url":"https://arxiv.org/abs/2410.06940v2","paper_title":"Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think","code":"https://github.com/sihyun-yu/REPA","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":2,"n_samples":9,"n_pointer_only_licence":2}},{"rank_in_archive_order":19,"model":"RAR-XXL, autoregressive","metrics":{"FID":"1.48"},"uses_additional_data":false,"paper_date":"2024-11-01","paper":"/paper/randomized-autoregressive-visual-generation","paper_url":"https://arxiv.org/abs/2411.00776v1","paper_title":"Randomized Autoregressive Visual Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":20,"model":"RAR-XL, autoregressive","metrics":{"FID":"1.50"},"uses_additional_data":false,"paper_date":"2024-11-01","paper":"/paper/randomized-autoregressive-visual-generation","paper_url":"https://arxiv.org/abs/2411.00776v1","paper_title":"Randomized Autoregressive Visual Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":21,"model":"MaskBit","metrics":{"FID":"1.52"},"uses_additional_data":false,"paper_date":"2024-09-24","paper":"/paper/maskbit-embedding-free-image-generation-via","paper_url":"https://arxiv.org/abs/2409.16211v2","paper_title":"MaskBit: Embedding-free Image Generation via Bit Tokens","code":"https://github.com/markweberdev/maskbit","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":1,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"GMem (w/o guidance)","metrics":{"FID":"1.53"},"uses_additional_data":false,"paper_date":"2024-12-11","paper":"/paper/generative-modeling-with-explicit-memory","paper_url":"https://arxiv.org/abs/2412.08781v1","paper_title":"Generative Modeling with Explicit Memory","code":"https://github.com/lins-lab/gmem","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":7,"n_samples":15,"n_pointer_only_licence":15}},{"rank_in_archive_order":23,"model":"ELM","metrics":{"FID":"1.54"},"uses_additional_data":false,"paper_date":"2024-10-21","paper":"/paper/elucidating-the-design-space-of-language","paper_url":"https://arxiv.org/abs/2410.16257v1","paper_title":"Elucidating the design space of language models for image generation","code":"https://github.com/Pepper-lll/LMforImageGeneration","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"MAR-H, Diff Loss","metrics":{"FID":"1.55"},"uses_additional_data":false,"paper_date":"2024-06-17","paper":"/paper/autoregressive-image-generation-without","paper_url":"https://arxiv.org/abs/2406.11838v3","paper_title":"Autoregressive Image Generation without Vector Quantization","code":"https://github.com/lth14/mar","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":5,"n_samples":13,"n_pointer_only_licence":1}},{"rank_in_archive_order":25,"model":"PaGoDA","metrics":{"FID":"1.56"},"uses_additional_data":false,"paper_date":"2024-05-23","paper":"/paper/pagoda-progressive-growing-of-a-one-step","paper_url":"https://arxiv.org/abs/2405.14822v2","paper_title":"PaGoDA: Progressive Growing of a One-Step Generator from a Low-Resolution Diffusion Teacher","code":"https://github.com/sony/pagoda","n_code_links":1,"syntology":{"n_ran":15,"n_unverified":2,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":26,"model":"ViT-XL/2 with limited Interval Guidance","metrics":{"FID":"1.57"},"uses_additional_data":false,"paper_date":"2023-03-16","paper":"/paper/efficient-diffusion-training-via-min-snr","paper_url":"https://arxiv.org/abs/2303.09556v3","paper_title":"Efficient Diffusion Training via Min-SNR Weighting Strategy","code":"https://github.com/tiankaihang/min-snr-diffusion-training","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":2}},{"rank_in_archive_order":27,"model":"MDTv2","metrics":{"FID":"1.58"},"uses_additional_data":false,"paper_date":"2023-03-25","paper":"/paper/masked-diffusion-transformer-is-a-strong","paper_url":"https://arxiv.org/abs/2303.14389v2","paper_title":"MDTv2: Masked Diffusion Transformer is a Strong Image Synthesizer","code":"https://github.com/sail-sg/mdt","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":28,"model":"SiT-XL + SRA","metrics":{"FID":"1.58"},"uses_additional_data":false,"paper_date":"2025-05-05","paper":"/paper/no-other-representation-component-is-needed","paper_url":"https://arxiv.org/abs/2505.02831v4","paper_title":"No Other Representation Component Is Needed: Diffusion Transformers Can Provide Representation Guidance by Themselves","code":"https://github.com/vvvvvjdy/sra","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":1,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":29,"model":"RobustTok-L","metrics":{"FID":"1.60"},"uses_additional_data":false,"paper_date":"2025-03-11","paper":"/paper/robust-latent-matters-boosting-image","paper_url":"https://arxiv.org/abs/2503.08354v1","paper_title":"Robust Latent Matters: Boosting Image Generation with Sampling Error","code":"https://github.com/lxa9867/imagefolder","n_code_links":1,"syntology":null},{"rank_in_archive_order":30,"model":"DiMR-G/2R","metrics":{"FID":"1.63"},"uses_additional_data":false,"paper_date":"2024-06-13","paper":"/paper/alleviating-distortion-in-image-generation","paper_url":"https://arxiv.org/abs/2406.09416v2","paper_title":"Alleviating Distortion in Image Generation via Multi-Resolution Diffusion Models and Time-Dependent Layer Normalization","code":"https://github.com/qihao067/DiMR","n_code_links":1,"syntology":{"n_ran":10,"n_unverified":6,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":31,"model":"FlowAR","metrics":{"FID":"1.65"},"uses_additional_data":false,"paper_date":"2024-12-19","paper":"/paper/flowar-scale-wise-autoregressive-image","paper_url":"https://arxiv.org/abs/2412.15205v1","paper_title":"FlowAR: Scale-wise Autoregressive Image Generation Meets Flow Matching","code":"https://github.com/oliverrensu/flowar","n_code_links":1,"syntology":null},{"rank_in_archive_order":32,"model":"FACM (1-step)","metrics":{"FID":"1.70","NFE":"1"},"uses_additional_data":false,"paper_date":"2025-07-04","paper":"/paper/flow-anchored-consistency-models","paper_url":"https://arxiv.org/abs/2507.03738v1","paper_title":"Flow-Anchored Consistency Models","code":"https://github.com/ali-vilab/FACM","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":6,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":33,"model":"DiT-XL/2 with CADS","metrics":{"FID":"1.70"},"uses_additional_data":false,"paper_date":"2023-10-26","paper":"/paper/cads-unleashing-the-diversity-of-diffusion","paper_url":"https://arxiv.org/abs/2310.17347v4","paper_title":"CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Sampling","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":34,"model":"DiMR-XL/2R","metrics":{"FID":"1.70"},"uses_additional_data":false,"paper_date":"2024-06-13","paper":"/paper/alleviating-distortion-in-image-generation","paper_url":"https://arxiv.org/abs/2406.09416v2","paper_title":"Alleviating Distortion in Image Generation via Multi-Resolution Diffusion Models and Time-Dependent Layer Normalization","code":"https://github.com/qihao067/DiMR","n_code_links":1,"syntology":{"n_ran":10,"n_unverified":6,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":35,"model":"RAR-L, autoregressive","metrics":{"FID":"1.70"},"uses_additional_data":false,"paper_date":"2024-11-01","paper":"/paper/randomized-autoregressive-visual-generation","paper_url":"https://arxiv.org/abs/2411.00776v1","paper_title":"Randomized Autoregressive Visual Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":36,"model":"DiffiT","metrics":{"FID":"1.73"},"uses_additional_data":false,"paper_date":"2023-12-04","paper":"/paper/diffit-diffusion-vision-transformers-for","paper_url":"https://arxiv.org/abs/2312.02139v3","paper_title":"DiffiT: Diffusion Vision Transformers for Image Generation","code":"https://github.com/nvlabs/diffit","n_code_links":1,"syntology":{"n_ran":18,"n_unverified":7,"n_samples":25,"n_pointer_only_licence":25}},{"rank_in_archive_order":37,"model":"VAR (Visual Autoregressive)","metrics":{"FID":"1.73"},"uses_additional_data":false,"paper_date":"2024-04-03","paper":"/paper/visual-autoregressive-modeling-scalable-image","paper_url":"https://arxiv.org/abs/2404.02905v2","paper_title":"Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction","code":"https://github.com/FoundationVision/VAR","n_code_links":3,"syntology":{"n_ran":5,"n_unverified":6,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":38,"model":"MAGVIT-v2","metrics":{"FID":"1.78"},"uses_additional_data":false,"paper_date":"2023-10-09","paper":"/paper/language-model-beats-diffusion-tokenizer-is","paper_url":"https://arxiv.org/abs/2310.05737v3","paper_title":"Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation","code":"https://github.com/jy0205/Pyramid-Flow","n_code_links":3,"syntology":{"n_ran":12,"n_unverified":8,"n_samples":20,"n_pointer_only_licence":0}},{"rank_in_archive_order":39,"model":"MAR-L, Diff Loss","metrics":{"FID":"1.78"},"uses_additional_data":false,"paper_date":"2024-06-17","paper":"/paper/autoregressive-image-generation-without","paper_url":"https://arxiv.org/abs/2406.11838v3","paper_title":"Autoregressive Image Generation without Vector Quantization","code":"https://github.com/lth14/mar","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":5,"n_samples":13,"n_pointer_only_licence":1}},{"rank_in_archive_order":40,"model":"MDT","metrics":{"FID":"1.79"},"uses_additional_data":false,"paper_date":"2023-03-25","paper":"/paper/masked-diffusion-transformer-is-a-strong","paper_url":"https://arxiv.org/abs/2303.14389v2","paper_title":"MDTv2: Masked Diffusion Transformer is a Strong Image Synthesizer","code":"https://github.com/sail-sg/mdt","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":41,"model":"Discriminator Guidance","metrics":{"FID":"1.83"},"uses_additional_data":false,"paper_date":"2022-11-28","paper":"/paper/refining-generative-process-with","paper_url":"https://arxiv.org/abs/2211.17091v4","paper_title":"Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models","code":"https://github.com/alsdudrla10/DG","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":42,"model":"DoD-XL","metrics":{"FID":"1.83"},"uses_additional_data":false,"paper_date":"2024-10-11","paper":"/paper/diffusion-models-need-visual-priors-for-image","paper_url":"https://arxiv.org/abs/2410.08531v1","paper_title":"Diffusion Models Need Visual Priors for Image Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":43,"model":"RobustTok-B","metrics":{"FID":"1.83"},"uses_additional_data":false,"paper_date":"2025-03-11","paper":"/paper/robust-latent-matters-boosting-image","paper_url":"https://arxiv.org/abs/2503.08354v1","paper_title":"Robust Latent Matters: Boosting Image Generation with Sampling Error","code":"https://github.com/lxa9867/imagefolder","n_code_links":1,"syntology":null},{"rank_in_archive_order":44,"model":"ARPG-XXL","metrics":{"FID":"1.94"},"uses_additional_data":false,"paper_date":"2025-03-13","paper":"/paper/autoregressive-image-generation-with","paper_url":"https://arxiv.org/abs/2503.10568v1","paper_title":"Autoregressive Image Generation with Randomized Parallel Decoding","code":"https://github.com/hp-l33/ARPG","n_code_links":1,"syntology":{"n_ran":11,"n_unverified":7,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":45,"model":"RAR-B, autoregressive","metrics":{"FID":"1.95"},"uses_additional_data":false,"paper_date":"2024-11-01","paper":"/paper/randomized-autoregressive-visual-generation","paper_url":"https://arxiv.org/abs/2411.00776v1","paper_title":"Randomized Autoregressive Visual Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":46,"model":"TiTok-S-128","metrics":{"FID":"1.97"},"uses_additional_data":false,"paper_date":"2024-06-11","paper":"/paper/an-image-is-worth-32-tokens-for","paper_url":"https://arxiv.org/abs/2406.07550v1","paper_title":"An Image is Worth 32 Tokens for Reconstruction and Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":2,"syntology":null},{"rank_in_archive_order":47,"model":"PixelFlow","metrics":{"FID":"1.98"},"uses_additional_data":false,"paper_date":"2025-04-10","paper":"/paper/pixelflow-pixel-space-generative-models-with","paper_url":"https://arxiv.org/abs/2504.07963v1","paper_title":"PixelFlow: Pixel-Space Generative Models with Flow","code":"https://github.com/shoufachen/pixelflow","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":5,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":48,"model":"RDM","metrics":{"FID":"1.99"},"uses_additional_data":false,"paper_date":"2023-09-04","paper":"/paper/relay-diffusion-unifying-diffusion-process-1","paper_url":"https://arxiv.org/abs/2309.03350v1","paper_title":"Relay Diffusion: Unifying diffusion process across resolutions for image synthesis","code":"https://github.com/THUDM/RelayDiffusion","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":5,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":49,"model":"FasterDiT-XL/2","metrics":{"FID":"2.03"},"uses_additional_data":false,"paper_date":"2024-10-14","paper":"/paper/fasterdit-towards-faster-diffusion","paper_url":"https://arxiv.org/abs/2410.10356v2","paper_title":"FasterDiT: Towards Faster Diffusion Transformers Training without Architecture Modification","code":"https://github.com/hustvl/LightningDiT","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":4,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":50,"model":"LEGO-XL","metrics":{"FID":"2.05","Inception score":"338.08"},"uses_additional_data":false,"paper_date":"2023-10-10","paper":"/paper/learning-stackable-and-skippable-lego-bricks","paper_url":"https://arxiv.org/abs/2310.06389v3","paper_title":"Learning Stackable and Skippable LEGO Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling","code":"https://github.com/JegZheng/LEGODiffusion","n_code_links":1,"syntology":{"n_ran":11,"n_unverified":6,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":51,"model":"ARPG-XL","metrics":{"FID":"2.1"},"uses_additional_data":false,"paper_date":"2025-03-13","paper":"/paper/autoregressive-image-generation-with","paper_url":"https://arxiv.org/abs/2503.10568v1","paper_title":"Autoregressive Image Generation with Randomized Parallel Decoding","code":"https://github.com/hp-l33/ARPG","n_code_links":1,"syntology":{"n_ran":11,"n_unverified":7,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":52,"model":"StyleSAN-XL","metrics":{"FID":"2.14"},"uses_additional_data":false,"paper_date":"2023-01-30","paper":"/paper/adversarially-slicing-generative-networks","paper_url":"https://arxiv.org/abs/2301.12811v4","paper_title":"SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer","code":"https://github.com/sony/san","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":53,"model":"LlamaGen","metrics":{"FID":"2.18"},"uses_additional_data":false,"paper_date":"2024-06-10","paper":"/paper/autoregressive-model-beats-diffusion-llama","paper_url":"https://arxiv.org/abs/2406.06525v1","paper_title":"Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation","code":"https://github.com/foundationvision/llamagen","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":9,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":54,"model":"DiT-XL/2","metrics":{"FID":"2.27"},"uses_additional_data":false,"paper_date":"2022-12-19","paper":"/paper/scalable-diffusion-models-with-transformers","paper_url":"https://arxiv.org/abs/2212.09748v2","paper_title":"Scalable Diffusion Models with Transformers","code":"https://github.com/huggingface/diffusers","n_code_links":13,"syntology":{"n_ran":5,"n_unverified":15,"n_samples":20,"n_pointer_only_licence":4}},{"rank_in_archive_order":55,"model":"StyleGAN-XL","metrics":{"FID":"2.30"},"uses_additional_data":false,"paper_date":"2022-02-01","paper":"/paper/stylegan-xl-scaling-stylegan-to-large-diverse","paper_url":"https://arxiv.org/abs/2202.00273v2","paper_title":"StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets","code":"https://github.com/autonomousvision/stylegan-xl","n_code_links":2,"syntology":{"n_ran":14,"n_unverified":5,"n_samples":19,"n_pointer_only_licence":0}},{"rank_in_archive_order":56,"model":"MAR-B, Diff Loss","metrics":{"FID":"2.31"},"uses_additional_data":false,"paper_date":"2024-06-17","paper":"/paper/autoregressive-image-generation-without","paper_url":"https://arxiv.org/abs/2406.11838v3","paper_title":"Autoregressive Image Generation without Vector Quantization","code":"https://github.com/lth14/mar","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":5,"n_samples":13,"n_pointer_only_licence":1}},{"rank_in_archive_order":57,"model":"Open-MAGVIT2-XL","metrics":{"FID":"2.33"},"uses_additional_data":false,"paper_date":"2024-09-06","paper":"/paper/open-magvit2-an-open-source-project-toward","paper_url":"https://arxiv.org/abs/2409.04410v2","paper_title":"Open-MAGVIT2: An Open-Source Project Toward Democratizing Auto-regressive Visual Generation","code":"https://github.com/tencentarc/open-magvit2","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":1}},{"rank_in_archive_order":58,"model":"ACDiT","metrics":{"FID":"2.37"},"uses_additional_data":false,"paper_date":"2024-12-10","paper":"/paper/acdit-interpolating-autoregressive","paper_url":"https://arxiv.org/abs/2412.07720v1","paper_title":"ACDiT: Interpolating Autoregressive Conditional Modeling and Diffusion Transformer","code":"https://github.com/thunlp/acdit","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":5,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":59,"model":"ARPG-L","metrics":{"FID":"2.44"},"uses_additional_data":false,"paper_date":"2025-03-13","paper":"/paper/autoregressive-image-generation-with","paper_url":"https://arxiv.org/abs/2503.10568v1","paper_title":"Autoregressive Image Generation with Randomized Parallel Decoding","code":"https://github.com/hp-l33/ARPG","n_code_links":1,"syntology":{"n_ran":11,"n_unverified":7,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":60,"model":"TiTok-B-64","metrics":{"FID":"2.48"},"uses_additional_data":false,"paper_date":"2024-06-11","paper":"/paper/an-image-is-worth-32-tokens-for","paper_url":"https://arxiv.org/abs/2406.07550v1","paper_title":"An Image is Worth 32 Tokens for Reconstruction and Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":2,"syntology":null},{"rank_in_archive_order":61,"model":"GIVT-Causal-L+A","metrics":{"FID":"2.59"},"uses_additional_data":false,"paper_date":"2023-12-04","paper":"/paper/givt-generative-infinite-vocabulary","paper_url":"https://arxiv.org/abs/2312.02116v4","paper_title":"GIVT: Generative Infinite-Vocabulary Transformers","code":"https://github.com/google-research/big_vision","n_code_links":2,"syntology":null},{"rank_in_archive_order":62,"model":"Patch Diffusion","metrics":{"FID":"2.74"},"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":63,"model":"TiTok-B-32","metrics":{"FID":"2.77"},"uses_additional_data":false,"paper_date":"2024-06-11","paper":"/paper/an-image-is-worth-32-tokens-for","paper_url":"https://arxiv.org/abs/2406.07550v1","paper_title":"An Image is Worth 32 Tokens for Reconstruction and Generation","code":"https://github.com/bytedance/1d-tokenizer","n_code_links":2,"syntology":null},{"rank_in_archive_order":64,"model":"DoD-B","metrics":{"FID":"2.79"},"uses_additional_data":false,"paper_date":"2024-10-11","paper":"/paper/diffusion-models-need-visual-priors-for-image","paper_url":"https://arxiv.org/abs/2410.08531v1","paper_title":"Diffusion Models Need Visual Priors for Image Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":65,"model":"Poly-INR","metrics":{"FID":"2.86"},"uses_additional_data":false,"paper_date":"2023-03-20","paper":"/paper/polynomial-implicit-neural-representations","paper_url":"https://arxiv.org/abs/2303.11424v1","paper_title":"Polynomial Implicit Neural Representations For Large Diverse Datasets","code":"https://github.com/rajhans0/poly_inr","n_code_links":1,"syntology":null},{"rank_in_archive_order":66,"model":"MGVQ","metrics":{"FID":"3.02","Inception score":"294.1"},"uses_additional_data":false,"paper_date":"2025-07-14","paper":"/paper/mgvq-could-vq-vae-beat-vae-a-generalizable","paper_url":"https://arxiv.org/abs/2507.07997","paper_title":"MGVQ: Could VQ-VAE Beat VAE? A Generalizable Tokenizer with Multi-group Quantization","code":"https://github.com/MKJia/MGVQ","n_code_links":1,"syntology":null},{"rank_in_archive_order":67,"model":"ADM-G++ (FID)","metrics":{"FID":"3.18"},"uses_additional_data":false,"paper_date":"2022-11-28","paper":"/paper/refining-generative-process-with","paper_url":"https://arxiv.org/abs/2211.17091v4","paper_title":"Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models","code":"https://github.com/alsdudrla10/DG","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":68,"model":"DiGIT-0.7B","metrics":{"FID":"3.39","Inception score":"205.96"},"uses_additional_data":false,"paper_date":"2024-10-16","paper":"/paper/stabilize-the-latent-space-for-image","paper_url":"https://arxiv.org/abs/2410.12490v2","paper_title":"Stabilize the Latent Space for Image Autoregressive Modeling: A Unified Perspective","code":"https://github.com/DAMO-NLP-SG/DiGIT","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":2,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":69,"model":"Contextual RQ-Transformer","metrics":{"FID":"3.41"},"uses_additional_data":false,"paper_date":"2022-06-09","paper":"/paper/draft-and-revise-effective-image-generation","paper_url":"https://arxiv.org/abs/2206.04452v1","paper_title":"Draft-and-Revise: Effective Image Generation with Contextual RQ-Transformer","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":70,"model":"GigaGAN","metrics":{"FID":"3.45"},"uses_additional_data":false,"paper_date":"2023-03-09","paper":"/paper/scaling-up-gans-for-text-to-image-synthesis","paper_url":"https://arxiv.org/abs/2303.05511v2","paper_title":"Scaling up GANs for Text-to-Image Synthesis","code":"https://github.com/lucidrains/gigagan-pytorch","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":6,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":71,"model":"RCG-L (w/o guidance)","metrics":{"FID":"3.49"},"uses_additional_data":false,"paper_date":"2023-12-06","paper":"/paper/self-conditioned-image-generation-via","paper_url":"https://arxiv.org/abs/2312.03701v4","paper_title":"Return of Unconditional Generation: A Self-supervised Representation Generation Method","code":"https://github.com/LTH14/rcg","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":6,"n_samples":18,"n_pointer_only_licence":0}},{"rank_in_archive_order":72,"model":"BIGRoC-gt (Guided-Diffusion)","metrics":{"FID":"3.63"},"uses_additional_data":false,"paper_date":"2021-08-08","paper":"/paper/bigroc-boosting-image-generation-via-a-robust","paper_url":"https://arxiv.org/abs/2108.03702v4","paper_title":"BIGRoC: Boosting Image Generation via a Robust Classifier","code":"https://github.com/royg27/BIGRoC","n_code_links":1,"syntology":null},{"rank_in_archive_order":73,"model":"MAGVIT-v2 (w/o guidance)","metrics":{"FID":"3.65"},"uses_additional_data":false,"paper_date":"2023-10-09","paper":"/paper/language-model-beats-diffusion-tokenizer-is","paper_url":"https://arxiv.org/abs/2310.05737v3","paper_title":"Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation","code":"https://github.com/jy0205/Pyramid-Flow","n_code_links":3,"syntology":{"n_ran":12,"n_unverified":8,"n_samples":20,"n_pointer_only_licence":0}},{"rank_in_archive_order":74,"model":"BIGRoC-pl (Guided-Diffusion)","metrics":{"FID":"3.69"},"uses_additional_data":false,"paper_date":"2021-08-08","paper":"/paper/bigroc-boosting-image-generation-via-a-robust","paper_url":"https://arxiv.org/abs/2108.03702v4","paper_title":"BIGRoC: Boosting Image Generation via a Robust Classifier","code":"https://github.com/royg27/BIGRoC","n_code_links":1,"syntology":null},{"rank_in_archive_order":75,"model":"simple diffusion (U-Net)","metrics":{"FID":"3.71"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/simple-diffusion-end-to-end-diffusion-for","paper_url":"https://arxiv.org/abs/2301.11093v2","paper_title":"Simple diffusion: End-to-end diffusion for high resolution images","code":"https://github.com/fashn-AI/tryondiffusion","n_code_links":1,"syntology":null},{"rank_in_archive_order":76,"model":"simple diffusion (U-ViT, L)","metrics":{"FID":"3.75"},"uses_additional_data":false,"paper_date":"2023-01-26","paper":"/paper/simple-diffusion-end-to-end-diffusion-for","paper_url":"https://arxiv.org/abs/2301.11093v2","paper_title":"Simple diffusion: End-to-end diffusion for high resolution images","code":"https://github.com/fashn-AI/tryondiffusion","n_code_links":1,"syntology":null},{"rank_in_archive_order":77,"model":"RQ-Transformer","metrics":{"FID":"3.83"},"uses_additional_data":false,"paper_date":"2022-03-03","paper":"/paper/autoregressive-image-generation-using","paper_url":"https://arxiv.org/abs/2203.01941v2","paper_title":"Autoregressive Image Generation using Residual Quantization","code":"https://github.com/kakaobrain/rq-vae-transformer","n_code_links":4,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":78,"model":"ADM-G, ADM-U","metrics":{"FID":"3.94"},"uses_additional_data":false,"paper_date":"2021-05-11","paper":"/paper/diffusion-models-beat-gans-on-image-synthesis","paper_url":"https://arxiv.org/abs/2105.05233v4","paper_title":"Diffusion Models Beat GANs on Image Synthesis","code":"https://github.com/openai/guided-diffusion","n_code_links":21,"syntology":{"n_ran":27,"n_unverified":23,"n_samples":50,"n_pointer_only_licence":16}},{"rank_in_archive_order":79,"model":"ADM-G + EDS (ED-DPM, classifier_scale=0.75)","metrics":{"FID":"3.96"},"uses_additional_data":false,"paper_date":"2022-06-23","paper":"/paper/entropy-driven-sampling-and-training-scheme","paper_url":"https://arxiv.org/abs/2206.11474v5","paper_title":"Entropy-driven Sampling and Training Scheme for Conditional Diffusion Generation","code":"https://github.com/ZGCTroy/ED-DPM","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":80,"model":"MaskGIT (a=0.05)","metrics":{"FID":"4.02"},"uses_additional_data":false,"paper_date":"2022-02-08","paper":"/paper/maskgit-masked-generative-image-transformer","paper_url":"https://arxiv.org/abs/2202.04200v1","paper_title":"MaskGIT: Masked Generative Image Transformer","code":"https://github.com/lucidrains/soundstorm-pytorch","n_code_links":9,"syntology":{"n_ran":14,"n_unverified":7,"n_samples":21,"n_pointer_only_licence":4}},{"rank_in_archive_order":81,"model":"ADM-G + EDS + ECT (ED-DPM, classifier_scale=1.0)","metrics":{"FID":"4.09"},"uses_additional_data":false,"paper_date":"2022-06-23","paper":"/paper/entropy-driven-sampling-and-training-scheme","paper_url":"https://arxiv.org/abs/2206.11474v5","paper_title":"Entropy-driven Sampling and Training Scheme for Conditional Diffusion Generation","code":"https://github.com/ZGCTroy/ED-DPM","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":82,"model":"LDM","metrics":{"FID":"4.29"},"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":83,"model":"ADM-G++ (Recall)","metrics":{"FID":"4.45"},"uses_additional_data":false,"paper_date":"2022-11-28","paper":"/paper/refining-generative-process-with","paper_url":"https://arxiv.org/abs/2211.17091v4","paper_title":"Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models","code":"https://github.com/alsdudrla10/DG","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":84,"model":"LFM","metrics":{"FID":"4.46"},"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":85,"model":"RIN","metrics":{"FID":"4.51"},"uses_additional_data":false,"paper_date":"2022-12-22","paper":"/paper/scalable-adaptive-computation-for-iterative","paper_url":"https://arxiv.org/abs/2212.11972v2","paper_title":"Scalable Adaptive Computation for Iterative Generation","code":"https://github.com/google-research/pix2seq","n_code_links":2,"syntology":{"n_ran":21,"n_unverified":8,"n_samples":29,"n_pointer_only_licence":1}},{"rank_in_archive_order":86,"model":"ADM-G","metrics":{"FID":"4.59"},"uses_additional_data":false,"paper_date":"2021-05-11","paper":"/paper/diffusion-models-beat-gans-on-image-synthesis","paper_url":"https://arxiv.org/abs/2105.05233v4","paper_title":"Diffusion Models Beat GANs on Image Synthesis","code":"https://github.com/openai/guided-diffusion","n_code_links":21,"syntology":{"n_ran":27,"n_unverified":23,"n_samples":50,"n_pointer_only_licence":16}},{"rank_in_archive_order":87,"model":"CDM","metrics":{"FID":"4.88"},"uses_additional_data":false,"paper_date":"2021-05-30","paper":"/paper/cascaded-diffusion-models-for-high-fidelity","paper_url":"https://arxiv.org/abs/2106.15282v3","paper_title":"Cascaded Diffusion Models for High Fidelity Image Generation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":88,"model":"VQGAN+Transformer (k=600, p=1.0, a=0.05)","metrics":{"FID":"5.2"},"uses_additional_data":false,"paper_date":"2020-12-17","paper":"/paper/taming-transformers-for-high-resolution-image","paper_url":"https://arxiv.org/abs/2012.09841v3","paper_title":"Taming Transformers for High-Resolution Image Synthesis","code":"https://github.com/CompVis/taming-transformers","n_code_links":13,"syntology":{"n_ran":6,"n_unverified":0,"n_samples":6,"n_pointer_only_licence":4}},{"rank_in_archive_order":89,"model":"MaskGIT","metrics":{"FID":"6.18"},"uses_additional_data":false,"paper_date":"2022-02-08","paper":"/paper/maskgit-masked-generative-image-transformer","paper_url":"https://arxiv.org/abs/2202.04200v1","paper_title":"MaskGIT: Masked Generative Image Transformer","code":"https://github.com/lucidrains/soundstorm-pytorch","n_code_links":9,"syntology":{"n_ran":14,"n_unverified":7,"n_samples":21,"n_pointer_only_licence":4}},{"rank_in_archive_order":90,"model":"VQGAN+Transformer (k=mixed, p=1.0, a=0.005)","metrics":{"FID":"6.59"},"uses_additional_data":false,"paper_date":"2020-12-17","paper":"/paper/taming-transformers-for-high-resolution-image","paper_url":"https://arxiv.org/abs/2012.09841v3","paper_title":"Taming Transformers for High-Resolution Image Synthesis","code":"https://github.com/CompVis/taming-transformers","n_code_links":13,"syntology":{"n_ran":6,"n_unverified":0,"n_samples":6,"n_pointer_only_licence":4}},{"rank_in_archive_order":91,"model":"Polarity-BigGAN","metrics":{"FID":"6.82"},"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":92,"model":"BigGAN-deep","metrics":{"FID":"8.1"},"uses_additional_data":false,"paper_date":"2018-09-28","paper":"/paper/large-scale-gan-training-for-high-fidelity","paper_url":"http://arxiv.org/abs/1809.11096v2","paper_title":"Large Scale GAN Training for High Fidelity Natural Image Synthesis","code":"https://github.com/ajbrock/BigGAN-PyTorch","n_code_links":35,"syntology":{"n_ran":14,"n_unverified":27,"n_samples":41,"n_pointer_only_licence":4}},{"rank_in_archive_order":93,"model":"ADM","metrics":{"FID":"11.84"},"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":94,"model":"Improved DDPM","metrics":{"FID":"12.3"},"uses_additional_data":false,"paper_date":"2021-02-18","paper":"/paper/improved-denoising-diffusion-probabilistic-1","paper_url":"https://arxiv.org/abs/2102.09672v1","paper_title":"Improved Denoising Diffusion Probabilistic Models","code":"https://github.com/neonbjb/tortoise-tts","n_code_links":18,"syntology":{"n_ran":9,"n_unverified":2,"n_samples":11,"n_pointer_only_licence":0}}],"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 6,264 of the 9,581 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":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+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":64,"rows_with_any_sample_ran":64,"distinct_papers_with_graph_line":42,"distinct_papers_with_any_sample_ran":42,"samples_over_distinct_papers":{"n_ran":340,"n_unverified":208,"n_samples":548,"n_pointer_only_licence":98,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":530,"n_unverified":310,"n_samples":840,"n_pointer_only_licence":157,"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"}}}