简体中文
EnglishWebsite review
This review may include AI-generated content and is not individually verified. Domain, certificate and other facts come from the observed data on this page.
What is This Person Does Not Exist?
This Person Does Not Exist is a free online generator that creates photorealistic faces of people who have never existed. Every image is synthesized by the StyleGAN3 model, so the person in the photo is not real—there is no model, no photo shoot and no release form behind it.
What you can generate
- Random Face — a mix of men and women in one batch.
- Female or Male — eight head-and-shoulders portraits of one gender.
- Full Body — four standing figures instead of head shots.
Each click returns eight images at 1024×1024, and you can filter the batch by race, age band (1–21, 21–35, 35–50, 50+) and emotion (happy or neutral). Changing a dropdown reloads the set immediately. Downloads are free at 512×512 or full HD, with no signup, no daily cap and no queue, and the tool runs in a browser on desktop or phone.
Who it is for
- Designers and developers filling avatars, profile placeholders and mockups without licensing a stock portrait.
- Writers and researchers who need a plausible face for a fictional character or a demo dataset.
- Anyone who wants a profile picture that cannot be traced back to a real person.
Trade-offs to keep in mind
The faces are synthetic, so fine details such as ears, teeth, jewellery and background edges can look slightly off on close inspection, and full-body shots are more prone to odd hands or feet. Because the people do not exist, the images cannot be used to imply a real endorsement or to mislead someone about who they are talking to. Also note that the site is a browser tool only—nothing is installed locally, and generated faces are not saved to an account.
Next step: if you need a specific look, open the tab that matches your subject, set the race, age and emotion dropdowns, then hit Reload until a face fits. For a quick comparison with the original single-face generator, see This Person Does Not Exist.
How do I generate a random AI face on this site?
Open the site and choose one of the four tabs: Random Face, Female, Male or Full Body. Then optionally set the race, age or emotion dropdowns, click a photo, and save it at 512x512 or full HD. Hit Reload for eight new faces; there is no signup and no daily cap.
What each tab gives you
- Random Face – mixes men and women.
- Female – eight photos of a woman who does not exist.
- Male – eight photos of a man who does not exist.
- Full Body – four standing people instead of head shots.
The filters are the useful part
Three dropdowns sit above the photos, and changing one reloads the batch immediately:
- Race: Asian, White, Latino Hispanic, Middle Eastern, Indian, Black
- Age: 1–21, 21–35, 35–50, 50+
- Emotion: happy or neutral
Most face generators leave you to reload until something fits. Here you can ask directly for, say, a happy Asian face or a Black face aged 21–35, which saves a lot of clicking when you need a specific look.
A practical workflow
Say you are building a mockup and need a profile photo for a fictional customer support agent. Pick Female, set age to 21–35, set emotion to happy, then reload until one has the framing you want. Click it and save the full-HD version. Because the faces come from StyleGAN3, the result looks realistic but nobody in the photo is real — there is no model to pay and no release to sign.
Two things to keep in mind. The tool is online only, so it works the same on a phone but needs a connection. And if you need whole characters rather than head shots, the Full Body tab is the better starting point.
If you want to compare output styles, Generated Photos and This Person Does Not Exist are the obvious alternatives, though neither offers the same race, age and emotion filtering in one click.
Can I filter AI faces by race, age, and emotion?
Yes. ThisPersonNotExist.org puts three dropdowns above the photo grid and reloads the eight faces as soon as you change one, so you can ask for a narrower slice than most random-face tools allow.
What each filter does
- Race — Asian, White, Latino Hispanic, Middle Eastern, Indian, Black
- Age — 1-21, 21-35, 35-50, 50+
- Emotion — happy or neutral
You can combine them, for example an Indian face, a happy Asian face, or a Black face aged 21-35. The filters apply to the Random Face, Male, Female and Full Body tabs, and the results are free to download at 512x512 or 1024x1024 HD with no signup.
How to use it in practice
Think about which attributes actually matter for your job, because each filter narrows the pool and the faces you get are still randomly drawn from it — you don't get to keep re-rolling one specific person, only a fresh set of eight.
- A mockup or prototype where the face is small on screen: skip the filters entirely. Any face works, and you'll get through the layout faster.
- An avatar or profile picture where the face is the whole image: set emotion to happy or neutral deliberately, since a mismatched expression reads oddly next to your copy.
- A demo video or product walkthrough that names a region or age group: set race and age together so the on-screen face doesn't contradict the script.
- Full-body shots for character or scene work: expect only four standing people per reload rather than eight head shots, so budget more clicks.
Trade-offs worth knowing
Filtering buys relevance at the cost of variety — a highly specific combination (say, Middle Eastern, 50+, happy) gives you a much smaller effective pool than an unfiltered reload, so you may cycle through several sets before finding one that fits. Age bands are coarse, and race categories are broad labels rather than fine-grained control, so treat them as a first pass, not a casting tool.
If you only need a single generic head shot, the fastest path is to leave all three dropdowns alone and click Reload until something looks right. If you need a specific demographic for a client-facing asset, set all three filters first, then save several candidates at full HD before deciding — the set changes on every reload, so what you didn't download is gone.
For a broader set of generators with different model styles and filter options, Generated Photos and This Person Does Not Exist are worth comparing alongside this one.
What is the difference between the Random Face, Male, Female, and Full Body tabs?
The four tabs control what kind of synthetic person you get, not how the underlying generation works.
- Random Face mixes men and women in the same batch of eight head-and-shoulders portraits.
- Female returns eight portraits of women.
- Male returns eight portraits of men.
- Full Body switches from head shots to four standing people, so you get fewer images per reload but see the whole figure.
The practical difference is framing and count. Random Face, Female and Male are interchangeable when you only need a head shot: same 1024x1024 output, eight per click, filters for race, age and emotion applied on top. Full Body is the odd one out because a standing figure needs more vertical space, so the batch drops to four and the crop changes. If you are building avatars or profile pictures, stay with the three portrait tabs and let the race, age and emotion dropdowns narrow the result. If you need a character for a mockup, poster or game placeholder where clothing and posture matter, Full Body is the only tab that shows them.
One caveat worth testing yourself: emotion options are listed as happy or neutral, so don't expect a wide expressive range, and the age bands are broad (1-21, 21-35, 35-50, 50+), which means a "21-35" face can look quite different from one reload to the next. Generate a few batches per tab before committing, since the eight images per click are independent draws rather than variations of one person.
For a concrete workflow: pick Full Body first to find a pose and outfit you like, note the general look, then switch to Male or Female to generate matching head shots for the same fictional person. Related tools such as Generated Photos offer searchable libraries if you need consistent identity across many images rather than one-off random draws.
Where can I use these free AI-generated photos without worrying about model releases?
You can use them almost anywhere a real model's likeness would otherwise be a problem, because there is no model to pay, no photographer's copyright to clear and no release form to sign. The faces are synthetic, generated with StyleGAN3, so nobody in the photo exists and no real person's publicity rights are involved.
Practical places people put them:
- Profile pictures and avatars for forums, game accounts, test users or placeholder personas.
- Design mockups — drop a face into a website layout, app screen or pitch deck instead of a stock photo of a real person.
- Marketing and content drafts where you need a believable human presence before real photography is arranged.
- Datasets and QA — sample faces for testing face-detection, cropping or layout behaviour.
- Presentations and teaching material that need neutral, non-identifiable people.
Where the trade-off sits
The absence of a release is the whole point, but it does not remove every risk. A synthetic face can still resemble a real person by coincidence, and using one to imply endorsement, to mislead, or in a sensitive context such as medical, political or dating material can create problems a release would never have solved. Treat these as generic faces, not as "your" model.
| Use case | Fits well | Watch out |
|---|---|---|
| Avatar or placeholder profile | Yes — no identity attached | Don't imply it's a real named person |
| Website or app mockup | Yes — quick, free, HD | Swap for licensed photography before launch if brand matters |
| Ad or campaign draft | Yes for concepting | Avoid implying endorsement or fake testimonials |
| Sensitive or deceptive contexts | No | Misrepresentation risk regardless of the source |
A concrete next step
Decide first whether the image needs to be someone or merely stand in for someone. If it stands in — a mockup avatar, a wireframe, a test account — generate a face, download it, and note internally that it is AI-generated so nobody later assumes it is a real model. If it must be someone specific, or the context could mislead viewers, this tool is the wrong source and you need a real, released model.
For adjacent options, This Person Does Not Exist is the well-known single-face generator, and Generated Photos offers a searchable library of synthetic people.
Do I need to sign up or pay to download the HD faces?
No. You can download the HD faces without signing up or paying. The site states downloads are free with no signup, and there is no daily cap or queue. Each click gives you 8 faces at 1024x1024 HD, with a 512x512 option if you want a smaller file.
How to get a download in practice
- Pick a tab: Random Face, Female, Male or Full Body.
- Optionally set race, age or emotion to narrow the results.
- Click a photo, then save it at 512x512 or full HD. Use Reload for 8 new faces.
What to check before relying on it
- Licence and usage terms: "Free to download" and "free to use commercially" are different claims. The page says there is no model to pay and no release to sign, but confirm the site's own terms if you plan to use a face in advertising, an app, or anything customer-facing.
- Realism vs. suitability: These are StyleGAN3 faces, so they look photographic but belong to no real person. That is useful for mockups, avatars and placeholder profiles, but a generated face is a poor fit where viewers expect a real, named individual.
- Filter limits: The race, age and emotion dropdowns cover broad categories, so treat them as a starting point rather than precise casting. Full Body returns 4 standing people instead of 8 head shots.
If you only need a handful of placeholder portraits for a design draft, the no-signup, no-payment route is the fastest path. If you need a consistent character across many images, generate a batch first and shortlist before you build around one face. For a comparable free option, see This Person Does Not Exist.
Namecheap
IONOS SE
User reviews (0)