Prompt Library

FLUX.1 Kontext Prompts for In-Context Image Editing

30 copy-paste prompts

30 tested, copy-paste editing instructions for Black Forest Labs FLUX.1 Kontext. Swap backgrounds, transfer styles, change outfits, add or remove objects, and keep your subject perfectly consistent.

In short: This page contains 30 copy-paste ready prompts, organized into 6 categories with a description and pro tip for each. The first 15 prompts are free instantly — no signup needed. Hand-curated and tested by the AI Academy team.

By Louis Corneloup · Founder, Techpresso

Last updated May 15, 2026·Hand-curated & tested by the AI Academy team

Background Swaps

5 prompts

Studio to Outdoor Golden Hour

1/30

Using the [UPLOADED IMAGE], replace the background with a warm golden-hour outdoor setting, soft blurred trees and a low sun behind the subject. Keep the subject's pose, face, hair, clothing, and lighting on the face exactly the same. Match the new ambient light to the subject so the edges blend naturally, no harsh cutout outline.

Moves your subject from a plain or studio backdrop into a natural golden-hour outdoor scene without altering them.

💡

Pro tip: Name the original light direction ('lit from the left') so Kontext relights the background to match and avoids a pasted-on look.

Clean White Seamless Studio

2/30

Take the [UPLOADED IMAGE] and replace the background with a clean seamless white studio sweep, soft even softbox lighting, subtle floor shadow under the subject. Preserve the subject's exact appearance, pose, and colors. Remove any distracting objects from the original background.

Drops your subject onto a pristine white e-commerce-style studio background.

💡

Pro tip: Add 'keep a soft contact shadow under the feet' so the subject still feels grounded instead of floating.

City Street at Night

3/30

In the [UPLOADED IMAGE], change the background to a moody city street at night with glowing neon signs and bokeh from car lights, wet reflective pavement. Keep the subject unchanged but add subtle cool rim light on their edges to match the neon environment. Maintain the original framing and crop.

Places the subject in a cinematic night-city scene with environment-matched rim lighting.

💡

Pro tip: Specify the dominant neon color ('magenta and cyan signs') to control the rim-light tint cast onto the subject.

Cozy Interior Cafe

4/30

Using the [UPLOADED IMAGE], swap the background for a cozy cafe interior with warm hanging lights, blurred shelves of mugs, and a wooden table edge in the foreground. Keep the subject's face, expression, and clothing identical. Match the warm interior color temperature across the whole frame.

Transforms the setting into a warm, lived-in cafe while leaving the subject untouched.

💡

Pro tip: Mention 'shallow depth of field, background softly blurred' so the subject stays the clear focal point.

Transparent / Plain Color Cutout

5/30

From the [UPLOADED IMAGE], remove the background entirely and replace it with a flat solid [BRAND COLOR] background. Keep the subject's exact edges, hair detail, and shadows clean. Do not alter the subject's pose, colors, or proportions in any way.

Isolates the subject on a flat brand-color background for banners, ads, or thumbnails.

💡

Pro tip: Swap [BRAND COLOR] for a light neutral gray if you plan to key it out again later in a design tool.

Prompts get you started. Tutorials level you up.

A growing library of 300+ hands-on AI tutorials. New tutorials added every week.

Start 7-Day Free Trial

Style Transfer

5 prompts

Watercolor Illustration

6/30

Restyle the [UPLOADED IMAGE] as a soft watercolor illustration with visible paper texture, gentle color bleeds, and loose hand-painted edges. Keep the composition, subject identity, and overall layout recognizable. Apply the style uniformly across subject and background.

Converts the whole photo into a cohesive watercolor painting while keeping it recognizable.

💡

Pro tip: Add 'preserve facial likeness' so portraits stay identifiable even under heavy stylization.

Pixar / 3D Animation Style

7/30

Transform the [UPLOADED IMAGE] into a polished 3D animated movie style, soft global illumination, slightly stylized proportions, glossy eyes, and clean rounded surfaces. Keep the subject's recognizable features, hair color, and outfit. Maintain the original pose and camera angle.

Reimagines your photo as a glossy 3D-animated character render.

💡

Pro tip: Keep stylization moderate by adding 'subtle, not exaggerated proportions' if you want the result to still resemble the real person.

Vintage 1970s Film Photo

8/30

Restyle the [UPLOADED IMAGE] to look like a 1970s film photograph: warm faded colors, slight grain, gentle light leaks in one corner, and soft contrast. Keep the subject, pose, and composition exactly the same. Do not add or remove any objects.

Applies an authentic vintage analog-film grade to the existing image.

💡

Pro tip: Specify the corner for the light leak ('top-right') so the effect feels intentional rather than random.

Comic Book / Ink Style

9/30

Convert the [UPLOADED IMAGE] into a bold comic-book ink style with strong black outlines, halftone dot shading, and flat saturated colors. Preserve the subject's pose, expression, and key features. Keep the original framing.

Turns the photo into a flat, high-contrast comic-book panel.

💡

Pro tip: Add 'limited 4-color palette' for a retro print look, or 'modern clean line art' for a sharper graphic-novel feel.

Match the Style of a Reference

10/30

Apply the artistic style, color palette, and lighting mood described as [DESCRIBE REFERENCE STYLE] to the [UPLOADED IMAGE]. Keep the subject's identity, pose, and composition unchanged while transferring only the visual style. Blend the look consistently across the entire frame.

Restyles your image toward a described aesthetic without changing what is in the scene.

💡

Pro tip: Describe the reference in concrete terms ('muted teal-and-orange, high contrast, hazy backlight') rather than naming an artist for more reliable results.

Object Add / Remove

5 prompts

Remove Unwanted Background People

11/30

In the [UPLOADED IMAGE], remove the people and clutter in the background while keeping the main subject in the foreground completely unchanged. Realistically reconstruct the background area behind them with matching textures, lighting, and perspective. Keep the subject's edges and shadows intact.

Cleans photobombers and clutter out of the background while preserving the main subject.

💡

Pro tip: If the fill looks repetitive, add 'fill with plausible continuation of the existing wall and floor' to guide the reconstruction.

Remove a Specific Object

12/30

From the [UPLOADED IMAGE], remove the [OBJECT TO REMOVE] and seamlessly fill the area with the surrounding background, matching texture, color, and lighting. Do not alter anything else in the image. Keep all other objects and the subject exactly as they are.

Deletes one named object and inpaints the gap so it looks like it was never there.

💡

Pro tip: Be specific about location ('the coffee cup on the left side of the table') when multiple similar objects exist.

Add a Held Product

13/30

In the [UPLOADED IMAGE], add the subject naturally holding a [PRODUCT] in their hand, with correct hand grip, realistic scale, and shadows that match the existing scene lighting. Keep the subject's face, pose, and the rest of the image unchanged. Make the product reflections consistent with the environment.

Places a product convincingly into the subject's hand with matched lighting and shadows.

💡

Pro tip: State the approximate size ('a phone-sized device') so Kontext scales the added object correctly to the hand.

Add Atmospheric Elements

14/30

Add subtle atmospheric elements to the [UPLOADED IMAGE]: light fog near the ground, a few floating dust particles in the light beams, and gentle volumetric light. Keep the subject and all objects unchanged. Integrate the effects so they respect the existing light direction.

Adds depth and mood with fog, particles, and light rays without touching the subject.

💡

Pro tip: Keep it believable by adding 'subtle, realistic amount' so the scene does not become hazy or overdone.

Add a Logo or Text Sign

15/30

In the [UPLOADED IMAGE], add a [TEXT OR LOGO] on the [SURFACE / WALL / SHIRT], conforming to the surface perspective, curvature, and lighting. Keep the rest of the image, including the subject, completely unchanged. Make the added element look printed onto the surface, not pasted on top.

Maps text or branding onto a surface in the scene with correct perspective.

💡

Pro tip: Keep added text short; long phrases are where AI image models most often produce garbled characters.

Outfit & Appearance Edits

5 prompts

Change Outfit Color

16/30

In the [UPLOADED IMAGE], change the color of the subject's [GARMENT] to [NEW COLOR], keeping the exact same fabric texture, folds, wrinkles, and lighting. Do not change the cut, fit, or any other part of the image. Keep the subject's face and pose identical.

Recolors a specific garment while preserving its texture and the rest of the photo.

💡

Pro tip: Name the garment precisely ('the long-sleeve shirt, not the jacket') so only the intended item changes color.

Swap to Formal Business Attire

17/30

Edit the [UPLOADED IMAGE] so the subject is wearing a tailored [COLOR] business suit appropriate for a professional headshot, with realistic fabric drape and natural fit to their body. Keep the subject's face, hairstyle, expression, pose, and background unchanged. Match the suit lighting to the existing scene.

Replaces casual clothing with professional business attire for headshots and LinkedIn photos.

💡

Pro tip: Add 'collar and shoulders fitted to the existing pose' to avoid floating or mismatched clothing edges.

Change Hairstyle

18/30

In the [UPLOADED IMAGE], change the subject's hairstyle to [NEW HAIRSTYLE], keeping their face, skin tone, and facial features exactly the same. Make the new hair sit naturally on the head with realistic volume and lighting that matches the scene. Leave clothing and background unchanged.

Tries a new hairstyle on the subject while keeping their face and the scene intact.

💡

Pro tip: Specify length and texture ('shoulder-length wavy') rather than just a name for more predictable results.

Add or Remove Glasses

19/30

In the [UPLOADED IMAGE], add a pair of [STYLE] glasses on the subject's face, correctly positioned on the nose and ears, with realistic lens reflections and a faint shadow on the cheeks. Keep the face, expression, and everything else unchanged. Match the frame finish to the lighting.

Adds (or, when reversed, removes) eyewear that sits naturally on the subject's face.

💡

Pro tip: To remove glasses instead, reword to 'remove the glasses and reconstruct the eyes and brow naturally.'

Seasonal Outfit Swap

20/30

Edit the [UPLOADED IMAGE] so the subject is dressed for [SEASON], for example a warm coat, scarf, and gloves for winter, with realistic layering and fit. Keep the subject's face, pose, and background unchanged. Match the clothing lighting and shadows to the existing scene.

Re-dresses the subject for a different season while keeping the original setting.

💡

Pro tip: If you also want the scene to feel seasonal, run a separate background-swap edit afterward rather than combining both at once.

Product & Commercial Edits

5 prompts

Product on Clean Marketing Background

21/30

Take the product in the [UPLOADED IMAGE] and place it on a clean gradient studio background in [BRAND COLOR], with a soft reflection beneath it and gentle directional light from the upper left. Keep the product's exact shape, label, colors, and proportions unchanged. Remove any background clutter.

Turns a plain product photo into a polished marketing hero shot.

💡

Pro tip: Always add 'keep the label text and logo perfectly sharp and unchanged' so branding is not distorted.

Product Lifestyle Context

22/30

Place the product from the [UPLOADED IMAGE] into a realistic lifestyle scene: [SETTING, e.g. a marble kitchen counter with morning light]. Keep the product's shape, label, and colors identical. Add natural contact shadows and reflections so it sits believably in the new environment.

Composites your product into an aspirational real-world setting for ads and social.

💡

Pro tip: Describe one clear surface and light source; vague scenes make the product's grounding shadow look wrong.

Add Packaging or Shadow

23/30

In the [UPLOADED IMAGE], add a realistic soft drop shadow and subtle floor reflection beneath the product to make it look professionally lit. Keep the product itself and its colors completely unchanged. Match the shadow direction to the existing highlight on the product.

Adds professional shadow and reflection to a flat-looking product cutout.

💡

Pro tip: Match shadow softness to the light ('soft diffused shadow' for studio light, 'sharper shadow' for direct sun).

Mockup on Surface

24/30

Apply the design from the [UPLOADED IMAGE] onto a [PRODUCT SURFACE, e.g. a ceramic mug / t-shirt / phone case], wrapping it to the object's curvature and perspective with realistic lighting and fabric or material texture. Keep the design's colors and proportions accurate. Show the product in a clean studio setting.

Maps a flat design onto a physical product surface as a realistic mockup.

💡

Pro tip: Mention the material ('matte ceramic' vs 'glossy plastic') so reflections and texture match the real product.

Color Variant Generation

25/30

From the [UPLOADED IMAGE], create a version where the product is [NEW COLOR / FINISH] instead of its current color, keeping the exact same shape, material, lighting, reflections, and background. Do not change the label or any text. Keep the result photoreal and consistent with the original lighting.

Generates an alternate color variant of a product from a single photo.

💡

Pro tip: For metallic or glossy finishes, add 'preserve realistic specular highlights' so the new color still reads as the right material.

Go from copy-pasting to actually mastering AI.

AI Academy: 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ other tools. New tutorials added every week.

Start Your Free Trial

Character Consistency

5 prompts

Same Character, New Pose

26/30

Using the character in the [UPLOADED IMAGE], generate the same person in a new pose: [DESCRIBE POSE]. Keep their face, hairstyle, skin tone, outfit, and proportions exactly consistent. Maintain the same lighting style and overall look so it reads as the same character.

Re-poses the exact same character while keeping their identity locked.

💡

Pro tip: Describe the pose in plain anatomical terms ('arms crossed, slight smile, turned three-quarters') for the most reliable consistency.

Same Character, New Scene

27/30

Place the character from the [UPLOADED IMAGE] into a new setting: [DESCRIBE SCENE]. Keep their face, hair, outfit, and proportions identical to the source image. Adjust only the environment and the lighting on the character to match the new scene.

Drops a consistent character into a brand-new environment for storyboards or series.

💡

Pro tip: Use one source image as the visual anchor across your whole sequence so every scene references the same identity.

Consistent Character Expression Sheet

28/30

From the character in the [UPLOADED IMAGE], create the same person with a [EXPRESSION, e.g. surprised / laughing / serious] expression. Keep the face structure, hairstyle, outfit, and lighting identical, changing only the facial expression. Maintain the same framing and head angle.

Produces expression variants of one character for emotion sheets or comics.

💡

Pro tip: Change one expression per generation; bundling multiple changes erodes the likeness.

Two Characters, One Scene

29/30

Combine the character from the [UPLOADED IMAGE] with a second character described as [DESCRIBE SECOND CHARACTER] into a single scene where they are [INTERACTION]. Keep the first character's face, outfit, and proportions exactly consistent with the source. Match lighting and scale between both characters.

Builds a two-character scene while preserving the identity of your reference character.

💡

Pro tip: Anchor only one character to an upload and describe the other in text to keep the composition controllable.

Age or Outfit Continuity

30/30

Using the character in the [UPLOADED IMAGE], show the same person now wearing [NEW OUTFIT] while keeping their face, hairstyle, body proportions, and identity perfectly consistent. Match the lighting and rendering style of the source image. Do not change any facial features.

Keeps a character recognizable across outfit changes for a consistent series or brand mascot.

💡

Pro tip: Lock identity by ending the prompt with 'this must read as the exact same person' to discourage subtle face drift.

Frequently Asked Questions

FLUX.1 Kontext is Black Forest Labs' in-context image editing model: instead of generating a brand-new image from scratch, it takes an existing image plus a text instruction and edits it while preserving the original subject. That makes it ideal for background swaps, style transfer, outfit changes, and keeping a character consistent across shots. The prompts on this page are written as editing instructions that reference your [UPLOADED IMAGE].

Reference the source image, state exactly what to change, and explicitly state what to keep unchanged (face, pose, colors, framing). Being clear about preservation is what stops Kontext from redrawing the whole image. Mentioning lighting direction and adding realistic shadows or reflections is what makes edits blend instead of looking pasted on.

Kontext was trained to preserve the identity of the subject in your source image, so the same face, outfit, and proportions can carry across new poses and scenes. For best results, anchor each generation to a single clear reference image and change only one thing at a time. Adding a line like 'this must read as the exact same person' helps reduce subtle face drift.

You can, but each additional change increases the risk of unintended edits and erodes subject consistency. For reliable results, make one focused edit per generation, for example swap the background first, then run a second pass to change the outfit. Chaining edits this way keeps each step predictable and easy to undo.

Yes, all 30 prompts are free to copy, paste, and adapt. Replace the [BRACKETED PLACEHOLDERS] with your own subject, product, colors, or scene, then upload your image and run them in any FLUX.1 Kontext interface. They work as written and are meant to be tweaked for your specific edits.

Prompts are the starting line. Tutorials are the finish.

A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.

Start 7-Day Free Trial Explore AI for Marketers

7-day free trial. Cancel anytime.

Browse All Prompt Collections