mlsanityData to 7 Oct 2026

Diffusion Models Beat GANs on Image Synthesis

arXiv 2105.05233 ↗First posted 11 May 2021cs.LG cs.AI cs.CVcited by 8,185

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

We show that diffusion models can achieve image sample quality superior to the current state-of-the-art generative models. We achieve this on unconditional image synthesis by finding a better architecture through a series of ablations. For conditional image synthesis, we further improve sample quality with classifier guidance: a simple, compute-efficient method for trading off diversity for fidelity using gradients from a classifier. We achieve an FID of 2.97 on ImageNet 128×128, 4.59 on ImageNet 256×256, and 7.72 on ImageNet 512×512, and we match BigGAN-deep even with as few as 25 forward passes per sample, all while maintaining better coverage of the distribution. Finally, we find that classifier guidance combines well with upsampling diffusion models, further improving FID to 3.94 on ImageNet 256×256 and 3.85 on ImageNet 512×512. We release our code at https://github.com/openai/guided-diffusion

Key facts. Cited by 8,185 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 61 of its 81 references that are in the map. First posted on arXiv 11 May 2021.

Builds on

61 of 81 references.

Closest ancestors · 6

Same direction, newest first

  1. 2021Improved Denoising Diffusion Probabilistic Models18 Feb
  2. 2021Knowledge Distillation in Iterative Generative Models for Improved Sampling Speed7 Jan
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  4. 2020Denoising Diffusion Implicit Models6 Oct
  5. 2020Denoising Diffusion Probabilistic Models19 Jun
  6. 2020Improved Techniques for Training Score-Based Generative Models16 Jun

Other references · 55

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2020Language Models are Few-Shot Learnerscited by 32,199
  5. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  6. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  7. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  8. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  9. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  10. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  11. 2013Intriguing properties of neural networkscited by 5,114
  12. 2017Neural Discrete Representation Learningcited by 4,650
38 more
  1. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  2. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  3. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  4. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  5. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  6. 2016Improved Techniques for Training GANscited by 2,938
  7. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  8. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  9. 2014Conditional Generative Adversarial Netscited by 2,152
  10. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  11. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  12. 2018Group Normalizationcited by 1,514
  13. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
  14. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  15. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  16. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,339
  17. 2017Mixed Precision Trainingcited by 1,135
  18. 2021StyleCLIP: Text-Driven Manipulation of StyleGAN Imagerycited by 990
  19. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 925
  20. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 635
  21. 2020WaveGrad: Estimating Gradients for Waveform Generationcited by 620
  22. 2016StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 575
  23. 2020Jukebox: A Generative Model for Musiccited by 535
  24. 2016RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentationcited by 463
  25. 2019Your Classifier is Secretly an Energy Based Model and You Should Treat it Like Onecited by 402
  26. 2016A Learned Representation For Artistic Stylecited by 340
  27. 2021Generating Images with Sparse Representationscited by 262
  28. 2020Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Imagescited by 261
  29. 2018cGANs with Projection Discriminatorcited by 253
  30. 2019Large Scale Adversarial Representation Learningcited by 238
  31. 2018A Note on the Inception Scorecited by 225
  32. 2016A Theory of Generative ConvNetcited by 174
  33. 2017Modulating early visual processing by languagecited by 167
  34. 2020Learning Energy-Based Models by Diffusion Recovery Likelihoodcited by 122
  35. 2016Neural Photo Editing with Introspective Adversarial Networkscited by 109
  36. 2020Adversarial score matching and improved sampling for image generationcited by 95
  37. 2021Generating images from caption and vice versa via CLIP-Guided Generative Latent Space Searchcited by 57
  38. 2019High-Fidelity Image Generation With Fewer Labelscited by 51

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 60; 2022: 485; 2023: 1792; 2024: 2264; 2025: 2232; 2026: 1352.

8,185 papers.

Sort:

By area:

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  3. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  4. 2022Scalable Diffusion Models with Transformerscited by 5,931
  5. 2022Classifier-Free Diffusion Guidancecited by 5,717
  6. 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,466
  7. 2022Flow Matching for Generative Modelingcited by 5,426
  8. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  9. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  10. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  11. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  12. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
38 more
  1. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  2. 2022InstructPix2Pix: Learning to Follow Image Editing Instructionscited by 2,591
  3. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,526
  4. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  5. 2022Video Diffusion Modelscited by 2,280
  6. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,231
  7. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Stepscited by 1,961
  8. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,903
  9. 2023Consistency Modelscited by 1,778
  10. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,765
  11. 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,755
  12. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,483
  13. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,465
  14. 2023IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Modelscited by 1,439
  15. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,419
  16. 2023AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuningcited by 1,418
  17. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,375
  18. 2021Palette: Image-to-Image Diffusion Modelscited by 1,296
  19. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,287
  20. 2023ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillationcited by 1,207
  21. 2022Imagic: Text-Based Real Image Editing with Diffusion Modelscited by 1,172
  22. 2022Null-text Inversion for Editing Real Images using Guided Diffusion Modelscited by 1,168
  23. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,155
  24. 2022Diffusion Models: A Comprehensive Survey of Methods and Applicationscited by 1,107
  25. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,099
  26. 2022MaskGIT: Masked Generative Image Transformercited by 1,082
  27. 2022Human Motion Diffusion Modelcited by 1,062
  28. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
  29. 2023ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generationcited by 1,010
  30. 2022Diffusion Models in Vision: A Surveycited by 1,008
  31. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,006
  32. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,005
  33. 2023One-step Diffusion with Distribution Matching Distillationcited by 978
  34. 2022Denoising Diffusion Restoration Modelscited by 951
  35. 2023Training Diffusion Models with Reinforcement Learningcited by 939
  36. 2021Blended Diffusion for Text-driven Editing of Natural Imagescited by 924
  37. 2022DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Modelscited by 924
  38. 2022eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoiserscited by 921

50 most cited of 8,185 All 8,185 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2023Conditional Image Generation with Pretrained Generative Model
  2. 2024Distilling Diffusion Models into Conditional GANs
  3. 2025Adaptive Destruction Processes for Diffusion Samplers
  4. 2023Adversarial Diffusion Distillation
  5. 2024Accelerating Video Diffusion Models via Distribution Matching
  6. 2025ADT: Tuning Diffusion Models with Adversarial Supervision
  7. 2025d-Sketch: Improving Visual Fidelity of Sketch-to-Image Translation with Pretrained Latent Diffusion Models without Retraining
  8. 2024Diffusion Models Need Visual Priors for Image Generation