Let’s explain how to create A chibi character using your face as references using the artificial intelligence. It is called chibi in the style inside the manga in which the characters are small, big heads and very adorable, a very particular style.
We are going to tell you a prompt that you can do, and we are going to teach you the result with Chatgpt and Gemini. The prompt will also help you for other AI Where you can create images using a reference photo. You just have to keep in mind that Each AI will have its own style When drawing this character, and that you can redo the result if you want to try again.
Create a Chibi doll with your factions
To create a chibi with your face you can do two things. Can Use a photo of yours as a reference Or simply describe your appearance, with the beautiful facial or the style of hair or eye color. If you use a photo all this data will be extracted from it, but the Chibi are never usually excessively realistic, so it is not so important.


Therefore, what you should do is upload your photo to the artificial intelligence chat you have chosen, and then Write a prompt with the petition to create the doll. We have used this Prompt:
“Create a chibi manga -style doll, similar to the reference image, with the person’s factions in attached photography. The doll must have a black shirt and jeans, blacks, with black shoes. The background must be that of a concert, and the doll is jumping by making the gesture of the heavy metal horns with the hand. The expression of the doll must be of joy.”
Here, in the prompt you must mention that you want the image factions. You can also Describe the clothes you want to wear In the event that the photo does not look good. In addition, you can describe a gesture you want to be doing, and whatever you want to appear at the bottom of the image.
Another thing to keep in mind is that you can specify the expression you want your drawing, as an expression of joy. Besides, You can add other things to the prompt such as the size or proportion of the image, or any other detail you want to add.


Result with Gemini
The result with Gemini It is quite good, although sometimes you may need to rebuild it a couple of times until you find something better. The style is not exactly the chibi that you can recognize in any manga, but it is quite approaching. The good thing is that Gemini will get more information about your photoas details of the clothes you wear even if you do not ask for them.


Result with chatgpt
The result with chatgpt is the bestand the style is much more cuqui and childish. You may not get so many details of the original photo, but it is possibly what you have in your head when you try to make one of these dolls.


Result with Grok
In both cases we have used the same promptalthough you can try to make some changes in the way of expressing yourself. You can use this prompt in other AI, such as Grok either Perplexity. The first one approaches Gemini, Per Ose is behind in terms of the execution of the gesture you ask for, and the second allows you to use third -party models.
In Xataka Basics | 14 Functions, tricks and ideas to squeeze the analysis of chatgpt images
GIPHY App Key not set. Please check settings