Wedding day makeup
She's getting married in June and hardly ever wears makeup. Before the trial with the makeup artist, she compares a light look and a dressier one to know which way to go.
Try on makeup on your photo with AI
Your portrait with natural makeup, then an evening look with bronze smoky eyes and matte red lips. To choose colours before you buy. Workflow visible here, no account needed.
Real workflow: a portrait of a dark-haired woman with no makeup in a white T-shirt, then the same woman in natural makeup and in an evening look with bronze smoky eyes and matte red lips. The face is retouched: smoother skin, stronger brows.
At a glance
- You provide
- 1 photo · a text or a brief
- What the workflow produces
- 2 images in 3:4
- Canvas structure
- 6 nodes wired by 4 connections
- Models wired in
- Flux Kontext
- Cost of one full run
- about 12 credits, roughly $0.14
- Account
- The demo is visible without an account; generating requires a free account, trial credits included
A lipstick swatched on the back of your hand says little about how it will look on your face, and an eyeshadow seen on an influencer says even less. This workflow starts from YOUR bare-faced portrait and shows two directions: natural everyday makeup, and a bolder evening look.
The hard part is applying makeup without transforming the face. Image models readily confuse makeup with retouching: smoothed skin, slimmer features, redrawn brows. The prompt describes each product and asks for the same face. Flux Kontext gets the colours right, but in the demo it also retouched the face, so read it as a colour idea.
To go further, try a different eye colour or a new hairstyle. Both looks cost twelve credits.
Actual renders from the demo
These are the files of the workflow above, exactly as the models produced them, unretouched. The starting point is shown when there is one.
Starting point
Render
Who it is for
Women weighing a wedding or evening look, makeup artists preparing a trial, small cosmetics brands.
Which files to provide
A bare-faced front-facing portrait in daylight, and each product described with its colour.
How it works
- 1
Start bare-faced
Front-facing, no makeup, in daylight. The model adds better than it replaces.
- 2
Describe product by product
Base, eyes, lashes, cheeks, lips: one product and one colour per area. "Evening makeup" alone leaves too much to the model.
- 3
Protect your features
Add "same eyebrows, same freckles, no retouching of the face". Without it, the model smooths and redraws.
- 4
Compare with the photo
Put the render next to the original. Keep the colours you like, even if the face has shifted a little.
Variations and use cases
The same canvas, tuned for different needs: each variation is two or three lines of prompt away.
The makeup artist and her client
The makeup artist sends two directions on the client's portrait before the appointment. The trial starts from a colour they picked together instead of a blank page.
The first red lipstick
She has never dared wear red. A matte red on her own photo, and she knows whether to buy it or stick with nude, without testing ten tubes in the shop.
The end-of-year party
A dressy evening is coming up. She tries bronze smoky eyes and a deep red on her own face to see whether the two together are too much, before buying the palette.
The cosmetics brand
A small brand shows its shades on portraits of volunteer customers for its social accounts, each render checked against the original photo before posting.
The actress and the character
The character wears 1980s makeup. The actress and the production's makeup artist see the look on her own face before the first trial in the makeup trailer.
Common mistakes
Writing just "evening makeup"
The model picks everything, and often too much. Name the eyeshadow, the liner, the blush and the lip colour.
Starting from a made-up face
The new look mixes with the old one and the colours blur. Take your makeup off for the photo.
Taking it as your face made up
In the demo, the model also retouched the features. Judge the colours, not the effect on your face shape.
Ignoring the light
A portrait under a yellow bulb throws every shade off. Take the photo by a window, in daylight.
Recommended models
Flux Kontext
Nano Banana 2.1
GPT Image 2.5 SunburstFrequently asked questions
Does my face change?
It shouldn't, but in the demo it did: a fuller face, thicker brows, freckles almost gone. In the evening look especially, the likeness drops.
Are the colours accurate?
As an idea, yes: nude and peach on one side, bronze and deep red on the other. An exact brand shade can't be guaranteed in a picture.
Can I keep my freckles?
The natural prompt asks for them, and the model still almost erased them. Insist in the prompt, and run it again if needed.
How much do the two looks cost?
Twelve credits, six per render on Flux Kontext. A third look on the same photo costs six more.
Why Flux Kontext?
Because it edits the image without rebuilding all of it, which keeps the hair, the T-shirt and the background. With makeup, it couldn't resist retouching the features: that is its limit on this template.
Can I try a specific look I saw online?
Yes, by describing it area by area. The model doesn't copy a reference photo; it follows the description.
Does it work on darker or olive skin?
Yes: describe shades that suit, a plum blush or a copper bronze for instance, and ask to keep the skin tone.
What output format?
The format of your photo, here 3:4, a vertical portrait like the ones phones take.
Try on makeup on your photo with AI
Use this workflowFree account, no card required. The workflow opens pre-filled in your canvas.