# Generated Faces in the Wild: Quantitative Comparison of Stable Diffusion, Midjourney and DALL-E 2

arXiv: https://arxiv.org/abs/2210.00586 · HTML: https://mlsanity.com/paper/2210.00586/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 11 of its 14 references that are in the map.
- First posted on arXiv 2 Oct 2022.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 11 of 11)

- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Pros and Cons of GAN Evaluation Measures](https://mlsanity.com/paper/1802.03446/) (2018)
- [Seeing What a GAN Cannot Generate](https://mlsanity.com/paper/1910.11626/) (2019)
- [Pros and Cons of GAN Evaluation Measures: New Developments](https://mlsanity.com/paper/2103.09396/) (2021)
- [Bias and Generalization in Deep Generative Models: An Empirical Study](https://mlsanity.com/paper/1811.03259/) (2018)
- [A very preliminary analysis of DALL-E 2](https://mlsanity.com/paper/2204.13807/) (2022)

## Built on by (most cited first, 30 of 55)

- [DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models](https://mlsanity.com/paper/2210.14896/) (2022)
- [AGIQA-3K: An Open Database for AI-Generated Image Quality Assessment](https://mlsanity.com/paper/2306.04717/) (2023)
- [What You See is What You Read? Improving Text-Image Alignment Evaluation](https://mlsanity.com/paper/2305.10400/) (2023)
- [Unlearning Concepts in Diffusion Model via Concept Domain Correction and Concept Preserving Gradient](https://mlsanity.com/paper/2405.15304/) (2024)
- [Efficient Diffusion Models for Vision: A Survey](https://mlsanity.com/paper/2210.09292/) (2022)
- [Combining Generative Artificial Intelligence (AI) and the Internet: Heading towards Evolution or Degradation?](https://mlsanity.com/paper/2303.01255/) (2023)
- [Diffusion Facial Forgery Detection](https://mlsanity.com/paper/2401.15859/) (2024)
- [LogGPT: Exploring ChatGPT for Log-Based Anomaly Detection](https://mlsanity.com/paper/2309.01189/) (2023)
- [Qualitative Failures of Image Generation Models and Their Application in Detecting Deepfakes](https://mlsanity.com/paper/2304.06470/) (2023)
- [WEDGE: A multi-weather autonomous driving dataset built from generative vision-language models](https://mlsanity.com/paper/2305.07528/) (2023)
- [WizMap: Scalable Interactive Visualization for Exploring Large Machine Learning Embeddings](https://mlsanity.com/paper/2306.09328/) (2023)
- [ADVEDM:Fine-grained Adversarial Attack against VLM-based Embodied Agents](https://mlsanity.com/paper/2509.16645/) (2025)
- [A Survey of Zero-Knowledge Proof Based Verifiable Machine Learning](https://mlsanity.com/paper/2502.18535/) (2025)
- [Second-Order Information Matters: Revisiting Machine Unlearning for Large Language Models](https://mlsanity.com/paper/2403.10557/) (2024)
- [FACTIFY3M: A Benchmark for Multimodal Fact Verification with Explainability through 5W Question-Answering](https://mlsanity.com/paper/2306.05523/) (2023)
- [Moderator: Moderating Text-to-Image Diffusion Models through Fine-grained Context-based Policies](https://mlsanity.com/paper/2408.07728/) (2024)
- [REALEDIT: Reddit Edits As a Large-scale Empirical Dataset for Image Transformations](https://mlsanity.com/paper/2502.03629/) (2025)
- [Rocks Coding, Not Development--A Human-Centric, Experimental Evaluation of LLM-Supported SE Tasks](https://mlsanity.com/paper/2402.05650/) (2024)
- [AI Art in Architecture](https://mlsanity.com/paper/2212.09399/) (2022)
- [GenFace: A Large-Scale Fine-Grained Face Forgery Benchmark and Cross Appearance-Edge Learning](https://mlsanity.com/paper/2402.02003/) (2024)
- [Using Diffusion Models to Generate Synthetic Labelled Data for Medical Image Segmentation](https://mlsanity.com/paper/2310.16794/) (2023)
- [Adaptive Mixed-Scale Feature Fusion Network for Blind AI-Generated Image Quality Assessment](https://mlsanity.com/paper/2404.15163/) (2024)
- [Evaluating a Synthetic Image Dataset Generated with Stable Diffusion](https://mlsanity.com/paper/2211.01777/) (2022)
- [Analyzing Quality, Bias, and Performance in Text-to-Image Generative Models](https://mlsanity.com/paper/2407.00138/) (2024)
- [Scamming the Scammers: Using ChatGPT to Reply Mails for Wasting Time and Resources](https://mlsanity.com/paper/2303.13521/) (2023)
- [Stylebreeder: Exploring and Democratizing Artistic Styles through Text-to-Image Models](https://mlsanity.com/paper/2406.14599/) (2024)
- [From Paper to Card: Transforming Design Implications with Generative AI](https://mlsanity.com/paper/2403.08137/) (2024)
- [Grimm in Wonderland: Prompt Engineering with Midjourney to Illustrate Fairytales](https://mlsanity.com/paper/2302.08961/) (2023)
- [SFLD: Reducing the content bias for AI-generated Image Detection](https://mlsanity.com/paper/2502.17105/) (2025)
- [A Misleading Gallery of Fluid Motion by Generative Artificial Intelligence](https://mlsanity.com/paper/2405.15406/) (2024)

## Abstract (arXiv)

The field of image synthesis has made great strides in the last couple of years. Recent models are capable of generating images with astonishing quality. Fine-grained evaluation of these models on some interesting categories such as faces is still missing. Here, we conduct a quantitative comparison of three popular systems including Stable Diffusion, Midjourney, and DALL-E 2 in their ability to generate photorealistic faces in the wild. We find that Stable Diffusion generates better faces than the other systems, according to the FID score. We also introduce a dataset of generated faces in the wild dubbed GFW, including a total of 15,076 faces. Furthermore, we hope that our study spurs follow-up research in assessing the generative models and improving them. Data and code are available at data and code, respectively.