Change your car's colour in a photo with AI

Your car photo with the body repainted in two colours, same car and same setting. To choose a wrap or a respray. Workflow visible here, no account needed.

The photo of the blue car, parked in the streetInteractive demo

Real workflow, real results. Pan and zoom freely.

Use this workflow

Real workflow: an old blue hatchback parked in a wet street, then the same car in metallic red and pearl white. Hubcaps, scratches and street do not move.

At a glance

You provide
1 photo · a text or a brief
What the workflow produces
2 images in 4:3
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 wrap or a respray costs a lot and is quickly regretted. This workflow starts from a phone photo of your car and shows it in two colours, in your street, with your wheels and your small marks, before signing the quote.

The hard part is that it has to be your car. A model that recomposes the image returns a brand new, perfect car on another background: pretty, and useless. Flux Kontext repaints while keeping the geometry pixel perfect: same body, same scratches, same street, only the colour changes.

The prompt specifies what keeps its colour, chrome, rubber, glass, otherwise the black bumper takes on the colour. To sell the car afterwards, see listing photos. The two colours 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

Your car photo

Render

The same car in bright metallic red, hubcaps and scratches unchangedGenerated with Flux Kontext
The same car in pearl white, in the same wet streetGenerated with Flux Kontext

Who it is for

Drivers, wrap installers, body shops and companies who want to see a colour on the real car before paying.

Which files to provide

A three quarter photo of the car in daylight, and the colours to test with their finish.

How it works

  1. 1

    Photograph the car from three quarters

    Front and side visible, in overcast daylight: reflections are soft and the colour reads. Full sun creates reflections that mislead about the colour.

  2. 2

    Name the colour and its finish

    Metallic red, pearl white, matte grey, glossy British racing green: the finish matters as much as the colour. It is what changes how reflections render.

  3. 3

    Say what keeps its colour

    Chrome, rubber, glass, headlights, plates. Without this list, black bumpers or tinted windows take on the new colour.

  4. 4

    Compare the details

    Hubcaps, scratches, mirrors: they prove it is your car. If one of them changed, rerun the image.

Variations and use cases

The same canvas, tuned for different needs: each variation is two or three lines of prompt away.

Before a wrap

Three film colours tested on the real car, in the real street, before paying several hundred euros: twelve credits for two colours.

The body shop or wrapper

The professional shows the client two proposals on a photo of their own car, during the quote, instead of a chart of films.

Restoring a classic

A collector car seen in its original colour or a period colour, before launching a full respray.

The company vehicle

A van or a company car tested in brand colours, before ordering the livery for the whole fleet.

The car for sale

Showing a buyer what a respray would look like, stating clearly that it is a simulation and not the current state.

The gift or the whim

Seeing your first car in the colour you always wanted, for fun, with no commitment.

Common mistakes

Using a model that recomposes

It returns a new car without your marks, on another background. The test only counts if it is your car: Flux Kontext keeps the geometry.

Forgetting the unpainted parts

Without a prompt, black bumpers, trims or windows take on the colour. List what keeps its colour.

Photographing in full sun

Hard reflections distort how the colour reads. An overcast sky gives soft reflections, like a showroom.

Presenting the simulation as the real state

In a sale listing, a car repainted in a photo must be flagged as a simulation. Otherwise it misleads the buyer.

Frequently asked questions

Can I do a two tone wrap?

Yes, by describing each area: black roof and mirrors, fir green body. The clearer the areas on the car, the cleaner the render.

What about changing the wheels?

Yes, in another prompt: Flux Kontext changes the described wheel while keeping the rest. Make one variant for the colour and another for the wheels.

Is the render true to the body shop's colour?

Not enough to pick a reference: a screen distorts colours. The render picks the colour family, the body shop's sample decides.

How much do the two colours cost?

Twelve credits, six per render on Flux Kontext. A third colour on the same photo costs six more credits.

Why Flux Kontext?

Because it repaints while keeping the geometry pixel perfect, as measured on façades: the car, the street and the framing stay those of your photo.

Does the car stay identical?

On the demo, yes: same hubcaps, same headlights, same scratches, same wet street. Only the body changes.

Can I test a matte finish?

Yes, by saying so: "satin matte grey, no glossy reflections". Matte shows mostly in the reflections, the prompt must describe them.

What about a motorbike or a bike?

The same prompt works: name the painted parts and those that keep their colour. The more small parts a vehicle has, the more the list matters.

Change your car's colour in a photo with AI

Use this workflow

Free account, no card required. The workflow opens pre-filled in your canvas.