This Person Does Not Exist - Random Face Generator
Philip Wang · Image
This Person Does Not Exist is a free AI face generator that shows a new synthetic portrait on every refresh. No sign-up, no app, just photorealistic faces of people who never existed.
About This Person Does Not Exist - Random Face Generator
What Is This Person Does Not Exist
This Person Does Not Exist is a single-page website that shows a photorealistic face of someone who never existed. Every refresh loads a brand new portrait. You can't search it, you can't pick a style, and there's no gallery to browse. It shows one face. Swap it the second you reload.
The site was built by Philip Wang, an engineer at Uber, and went live in February 2019. Under the hood it runs StyleGAN, the face generation model NVIDIA published in late 2018. The model learned what faces look like from a large set of real photographs, mostly high-resolution portraits, and then draws new ones that never belonged to anyone. Wang's point was simple: most people had no idea how convincing AI faces had become, so he put one in front of them.
The catch is that it does one thing and stops there. You can't hold a face, adjust it, or generate ten at once from a control panel. Save the image or lose it, because each reload replaces the last face for good. There's also no API, so it's not built for developers or pipelines.
Getting Started
- Open thispersondoesnotexist.com in any browser.
- Wait a moment while the site loads one face image.
- Refresh the page for a new face. Every reload shows a different person.
- Right-click and save the image if you want to keep that face, since refreshing discards it.
- Check the licensing before you reuse a saved portrait, because the terms sit with NVIDIA's model and the training data.
Product Information
A quick look at This Person Does Not Exist - Random Face Generator's pricing, supported platforms, and performance.
Best for
The users, tasks, and scenarios where this tool fits best.
Users
- Designers who need a quick placeholder face
- Researchers and students studying AI output
- Developers testing face-related pipelines
Tasks
- Filling a mockup with a placeholder photo
- Building a privacy-safe test set
- Demonstrating how far AI has come
- Explaining GANs in plain terms
Scenarios
- Teaching a class about AI images
- A quick design sprint
- Comparing AI face generators
Key features
A New Face on Every Refresh
The page shows one face at a time, and reloading swaps in a different person. There's no queue and no history, so the face in front of you is the only one you get. Save it before you refresh, or it's gone.
Photorealistic Synthetic Portraits
The faces come from StyleGAN, a model NVIDIA published in late 2018. It learned from a large set of real photographs and draws AI-generated people who never existed. Up close the portraits hold up well enough that viewers often can't tell they're fake. The underlying technique is NVIDIA StyleGAN2, the second version of the model NVIDIA shipped.
Zero Setup and No Account
There's nothing to install and nothing to sign up for. Open the URL and a face loads. That simplicity is the point, and it's why the site spread so fast when it launched in February 2019.
Runs Entirely in the Browser
Any modern browser on desktop or mobile can load the page. You don't run the model yourself or download any weights, so there's nothing to configure and nothing to maintain.
A Single-Purpose Page
The site does exactly one thing. No filters, no batch generation, no download button beyond saving the image. So it works as a plain fake face generator and nothing more. That narrow scope keeps it fast, but it also means it won't fit any workflow that needs control.
Pros and cons
Pros
- Works instantly with no account, install, or payment, so anyone can see a synthetic face in seconds.
- The portraits are convincing enough to make the point that AI faces are hard to spot.
- Generated faces belong to nobody, which makes them a safe stand-in for test data and mockups.
- Runs on any browser and needs no hardware or setup of your own.
Cons
- No controls at all, so you can't set age, gender, expression, or background, and you just refresh until something fits.
- No API, which rules it out for anyone who wants to generate faces at scale inside a product or pipeline.
- You lose each face the moment you reload, so saving is manual and easy to forget.
- Licensing and reuse terms aren't spelled out on the page, so you have to check NVIDIA's model terms before publishing a portrait.
Frequently asked questions
It's a free website that shows a photorealistic face of a person who never existed. Every refresh generates a new, unique portrait, and there's nothing to install or sign up for.
Related content
Explore related tools, skills, and articles for This Person Does Not Exist - Random Face Generator.
This Person Does Not Exist - Random Face Generator Alternatives
Google EmbeddingGemma 2
Google DeepMind · Writing · Image · CodingGoogle EmbeddingGemma 2 is a 740M-parameter open model that turns text, code, images, audio, and video frames into vectors in one 768-dimension space, running on phones and laptops.
Tractionwave
TractionWave · Image · MarketingTractionWave is an AI attention heatmap tool for static ads. Upload a PNG or JPG and it returns predicted attention, ranked hotspots, and fixes before you spend budget.
Imejis.io
Imejis.io · Image · MarketingImejis.io is an AI design studio for agents and marketing teams. Describe an image in plain words, and it writes the copy, builds the layout, reflows it to every size, then exports a finished file.