The anime part is not a separate cast of drawn characters, it is the style you ask for when you want a picture. This is what that request looks like inside a conversation, which of the companions take to it, what five free messages a month will not cover, and where the drawn style stops working.
You want the picture drawn rather than photographed. On Private Crush that is a sentence you type in the middle of a conversation, not a setting you find in a menu, and the image comes back into the same thread a few seconds later. This page follows that request from the message to the picture, and it is straight about the part most pages skip, which is what the anime style does not cover.
Somebody searching for an anime girlfriend usually wants one of two things, and it is worth splitting them first.
The first is a drawn character to talk to, someone out of a series, who behaves the way an anime character behaves. That is not what this is. There are 125 companions here and every one of them is written as a person: Yuki Tanaka is twenty two, streams games from Tokyo and works shifts in a manga shop, Sophie Anderson is twenty and studies animation in Toronto, Giulia Rossi makes costumes for a living in Rome. They talk like people with jobs, not like characters with catchphrases.
The second is the pictures. Somebody who says anime girlfriend often means they are tired of the same photoreal selfie and want the image in the style they like looking at. That one the product does, because images are generated inside the conversation and the style is part of what you ask for.
This page is about the second. If you want the category itself, written as a product page, the AI waifu generator page covers the same ground from the term side. Here the question is narrower: you are in a chat, you want a drawn picture, what happens.
Nothing opens and nothing loads in a side panel. You type the request as you would ask a person for a photo, you put the style in the sentence, and the picture appears in the thread underneath it.
Most people spend their first attempt learning that the style has to be said out loud. A message that only asks for a picture gets a photograph, because that is the default the companions are written with. Ask for it drawn in anime style with soft cel shading and the result changes completely.
The second thing people learn is that the scene matters more than the adjective. Anime style at her desk at two in the morning with the monitor light on her face produces something. Anime style on its own produces a portrait against nothing, the picture nobody keeps.
By the second or third session the requests get shorter and better. You stop describing the art and start describing the moment, because the style is one clause you keep and the rest is what changes. People settle into two or three pictures across a conversation rather than a run of ten, because an image arriving in the middle of something reads differently from a gallery.
The other thing the first week shows you is the free ceiling. Five messages a month without paying is enough to find out whether you like how she writes and to get a picture back. It is not enough to learn the style, which takes a handful of attempts.
This is the part that surprises people, and it is why the drawn pictures are worth anything at all.
Every companion carries a written appearance: colouring, hair, height, the way she does her eyes. That description goes into every image she sends, so the anime version is recognisably the same woman as the photoreal one, not a new character each time. Yuki has black hair with a purple tint and cat eye glasses in both. Mara Lupescu, one of the 23 premium companions, is tall with jet black hair and crimson underlights in both.
It also means the style sits on top of a person rather than replacing her with a genre. Sophie draws in the style herself. Giulia has spent years making costumes of characters. The cast is not Japanese by default either: 20 nationalities have two or more companions, and the spread runs 13 American, 6 Brazilian, 6 Colombian, 6 Spanish, 4 Japanese, 4 French, 4 Russian, 4 Venezuelan, 4 Czech, then a long tail of ones and twos. If a background matters to you more than the drawing does, starting from the companion is the better order.
The two requests behave differently, and they are easier to see next to each other.
| What you notice | Asking for a photo | Asking for anime style |
|---|---|---|
| What you have to type | The scene is enough, the look is already there | The style has to be in the message, every single time |
| How close it stays to her | Very close, since her description does the work | Looser, though the hair and colouring carry over |
| What it handles well | Rooms, clothes, ordinary evenings, hands usually | Expression, mood, anything stylised or lit strangely |
| Where it falls apart | Fingers, small text on objects, mirrors | Proportions drift and detailed backgrounds go soft |
| How long it takes | Seconds, inside the thread | The same, there is no separate queue for it |
The row worth reading twice is the fourth. Drawn images forgive what photographs get caught on, then get caught on different things. If images rather than the conversation are why you are here, asking for images inside the chat covers that side properly.
The difference between a picture you keep and one you scroll past is almost always in the sentence you typed. Five things, roughly in order of how much they matter.
The style is a request, not a setting. If you do not say it in the message you will get a photograph, and saying it again costs you nothing.
There is a version of this page that would let you believe more than is true. Here is what is missing.
One more sits under the rest. Images are generated fresh each time from a written description, so consistency is good rather than perfect, and the drawn ones vary more. Anybody expecting the same face down to the eyelashes across twenty pictures will end up counting flaws.
Yes, and it happens inside the conversation rather than in a separate tool. You ask for a picture and say the style in the same message, and the image is generated and appears in the thread a few seconds later, in the context of whatever you were talking about. Both realistic and anime looks work this way. There is no menu of styles to click through, so what comes back depends a good deal on how clearly you asked for it.
No, and it is the thing most people expect to find. There is no style selector, no preset list and nothing that stays switched on between pictures. Every image request is read as a message, so the style has to be part of the sentence each time. In practice you keep two or three words, anime style or cel shaded, at the front of the request and change the scene after them. It is more work than a toggle and rather more flexible.
Broadly yes. Every companion carries a written description of her face, hair, colouring and build, and that description goes into each image, which keeps a drawn picture recognisable as the same woman as a photographed one. It is consistency rather than a photograph of a real person, so details shift between images, and the drawn ones drift more than the photoreal ones.
No. All 125 are written as people, with a history, a job and a way of talking, and none of them is a character from a series or a drawing pretending to be a person. Several have anime somewhere in their lives, which is a different thing: Yuki Tanaka streams games from Tokyo and works in a manga shop, Sophie Anderson studies animation in Toronto, Giulia Rossi makes costumes in Rome. If you were hoping to talk to a character from something you watched, that is not what this is.
Five messages a month, which is the honest number rather than the flattering one. That is enough to see how a companion writes and to get a picture or two back, and not enough to learn how to ask for a style, which takes a handful of tries. Treat the free tier as a look at the product rather than a way to use it. Nothing is deleted when you run out, the count refills the following month.
Twenty dollars a month, twenty four dollars for a quarter, or sixty dollars for a year, which is five dollars a month billed yearly. Messages are unlimited on the paid plan, and images are generated inside the conversation the same way they are on the free one, so there is no separate credit balance and no token meter. One plan lifts the message ceiling, and that is the whole of what is sold here.
Yes. All 125 companions allow adult conversation and send adult content, and the style applies there the same way it does anywhere else. The usual advice holds and matters more there: say the style, describe the scene rather than the art, and change one element at a time. Everyone here is written as an adult, the site is for adults only, and nothing on it will produce an image of a minor.
Inside the same conversation, yes, which is why it is worth keeping one thread going instead of starting fresh each visit. Across separate conversations, no. A new conversation begins without what the old one held, so the style has to be stated again there, and no profile is being assembled about you in between. This is the most common surprise on the site rather than something a setting turns on.
No, neither exists. Private Crush is text with images generated in the conversation, and there is no video generation of any kind, no animated pictures, no voice notes, no voice calls and no video calls. No plan adds them and nothing here is a preview of them. If a moving picture or a voice is what you came for, it is better to know now than to pay twenty dollars and find out.
No. Conversations and the images inside them are stored against your account and only your account can read them, so they are not published anywhere, not shown to other users and not collected into a feed, because there is no feed. What is kept and for how long is set out in the privacy policy, and the account page has a delete option for when you want it gone.
If you want the drawn version of somebody rather than a drawn character, the order that works is to pick the person first and add the style second. Have a look and meet the companions.
