ChatGPT Image Commands for Effects: 2026 Guide

There is no secret list of magic slash commands for ChatGPT image generation. In ChatGPT, the command is the prompt: a plain-language instruction that describes the image, the effect, the constraints, and the format you want.

That is good news for business users, because the most reliable image commands are reusable sentence patterns. OpenAI's own guidance says effective prompts should identify the purpose, subject, action, setting, visual style, and any important lighting, framing, or constraints. For edits, it also recommends being explicit about what should change and what should stay the same.

This guide gives you a practical library of copy-paste prompt commands and shortcut-style commands for image effects in ChatGPT. A shortcut such as /splashhero is best treated as a reusable shorthand your team defines, saves, and expands into a full prompt. Use these shortcuts for campaign mockups, blog graphics, product concepts, thumbnails, internal presentations, moodboards, and social media visuals. Treat the examples as starting points, then replace the bracketed details with your own subject, brand, product, audience, and format.

Quick Start ChatGPT Image Commands

These are the commands most users should learn first. They are not official ChatGPT syntax. They are reliable prompt patterns that match how ChatGPT Images works: describe the image or upload an image and describe the change.

Goal	Command to paste into ChatGPT
Create a new image	Create a [format] image of [subject] in [setting], with [style/effect], [lighting], and [mood]. Use [aspect ratio]. Avoid [constraints].
Edit an uploaded image	Edit this image by changing only [specific element] to [new detail]. Keep the composition, lighting, perspective, and all other details the same.
Generate variations	Create three visual directions for this concept: one minimal, one premium, and one playful. Keep the same subject and aspect ratio.
Apply a style reference	Use Image 1 as the subject and Image 2 as the style reference. Keep the layout and objects from Image 1, but apply the colours, texture, and illustration approach from Image 2.
Add text	Add the headline "[SHORT TEXT]" in bold sans-serif type at the [position]. Use [colour]. Keep the text sharp and do not add any other words.
Transparent background	Create a clean cutout of [subject] on a transparent background. Keep crisp edges, natural shadows, and no extra objects.
Change aspect ratio	Regenerate this as a [16:9 / 1:1 / 9:16] image. Preserve the same subject, style, colours, and key details.
Make a business visual	Create a professional [blog hero / social ad / website banner / presentation image] for [business topic]. Use a clear visual metaphor, modern Australian business context, and no fake brand logos.

Shortcut-Style Commands You Can Save

If you want commands that feel like /splashhero, create your own shortcut library. The slash name is the easy-to-remember trigger; the expanded prompt is what you paste into ChatGPT, save in a team prompt document, or use inside a custom GPT's instructions.

These are not official universal ChatGPT slash commands. They are practical prompt shortcuts that make repeatable image generation faster for marketing, website, ecommerce, and content teams.

Shortcut	Use it for	Expanded command
/splashhero	Website or blog hero visuals	Create a 16:9 splash hero image for [topic/service/product]. Show [main subject] in a premium, high-impact composition with clean negative space for a headline overlay. Use cinematic lighting, strong depth, realistic detail, and no embedded text or fake logos.
/productcutout	Product images with transparent backgrounds	Create a clean transparent-background cutout of [product]. Keep the product shape, colour, labels, material, and proportions accurate. Use crisp edges, a natural soft shadow, and no extra objects.
/brandclean	Generic brand-safe visuals	Create a generic, ownable business visual for [topic]. Use modern professional styling, original shapes, no recognisable brand identity, no trademarked logos, no copied artwork, and no fake interface text.
/neonposter	Bold posters and social graphics	Create a bold poster image for [message/event/topic] with neon rim lighting, high contrast, dramatic shadows, a clear central subject, and space for short headline text. Use a [1:1 / 4:5 / 9:16] layout.
/flatdiagram	Explainers and process graphics	Create a flat vector-style diagram explaining [process/system]. Use 4-6 clear steps, simple icons, large readable labels, a clean 16:9 layout, and no tiny text.
/isomap	Systems, workflows, and architecture	Create an isometric map of [workflow/system/platform]. Show the main parts as connected zones with clear spacing, subtle shadows, professional technology styling, and only short readable labels.
/photorealedit	Realistic edits to uploaded images	Edit this image by changing only [specific element] to [new detail]. Keep the subject, camera angle, lighting direction, shadows, perspective, background realism, and all other details unchanged.
/backgroundswap	Replacing photo backgrounds	Replace the background with [new background]. Keep the subject's pose, proportions, edge detail, clothing/product details, and lighting direction realistic. Match the new background perspective and shadows.
/socialsquare	Square social media assets	Create a 1:1 social media image for [message]. Use a simple visual metaphor, strong focal point, bold colour contrast, mobile-readable composition, and no small text.
/storyvertical	Stories, reels covers, and vertical ads	Create a 9:16 vertical image for [campaign/message]. Place the main subject in the centre third, keep clean safe space at the top and bottom, use energetic lighting, and avoid clutter.
/textposter	Images with a short headline	Create a poster image with exactly this headline: "[SHORT HEADLINE]". Put it in large bold sans-serif type at [position]. Keep the text sharp, add no other words, and use [visual style].
/variation3	Exploring creative directions	Create three distinct versions of this image concept: one minimal, one premium, and one playful. Keep the same subject, aspect ratio, and business message, but vary the colour, lighting, and composition.

How to use your own shortcuts

1. Write the shortcut name first, such as /splashhero.
2. Add the missing details: topic=[...], subject=[...], audience=[...], style=[...].
3. Paste the expanded command into ChatGPT, or keep the shortcut and full prompt in a shared prompt library.
4. For a custom GPT, add the shortcut list to its instructions and tell it to expand each shortcut into the matching image prompt.

The Full ChatGPT Image Effect Command Library

Use these commands as practical templates. Replace the bracketed parts and keep the final sentence if you want more predictable results.

Photography and Camera Effects

Effect	Command
Cinematic photo	Create a cinematic photograph of [subject] in [setting], with dramatic but realistic lighting, shallow depth of field, natural colours, and a 16:9 composition. Avoid fake text or logos.
Editorial portrait	Create an editorial portrait of [person/role], looking [mood], in [environment]. Use soft directional light, realistic skin texture, and a clean magazine-style composition.
Product photography	Create a premium studio product photo of [product] on [surface/background], with softbox lighting, realistic shadows, crisp edges, and commercial ecommerce styling.
Macro detail	Create a macro photograph of [small detail], showing fine texture, shallow focus, realistic highlights, and a clean background.
Flat lay	Create an overhead flat lay of [objects] arranged neatly on [surface], with balanced spacing, soft shadows, and a clean editorial style.
Documentary realism	Create a documentary-style photo of [scene], natural light, candid composition, realistic imperfections, and no staged advertising feel.
Luxury campaign	Create a luxury campaign image for [product/service], minimal composition, refined materials, restrained colour palette, elegant lighting, and generous negative space.
Action freeze	Create a high-speed action photo of [subject] doing [action], frozen motion, crisp detail, dynamic angle, and realistic environmental blur.
Long exposure	Create a long-exposure photograph of [scene], with smooth motion trails, stable background detail, and realistic night or low-light atmosphere.
Wide-angle interior	Create a wide-angle interior photo of [room/workspace], natural perspective, balanced lighting, realistic materials, and no distorted objects.

Lighting and Atmosphere Effects

Effect	Command
Golden hour	Create an image of [subject] at golden hour, warm low sunlight from the side, long soft shadows, natural highlights, and a calm optimistic mood.
Soft window light	Create a realistic image of [subject] lit by soft natural window light from the left, gentle contrast, neutral background, and authentic details.
Neon rim light	Create a modern image of [subject] with neon rim lighting, dark environment, subtle reflections, high contrast, and clean futuristic mood.
High-key studio	Create a high-key studio image of [subject], bright white background, soft even light, minimal shadows, and crisp commercial presentation.
Low-key dramatic	Create a low-key dramatic image of [subject], deep shadows, focused highlight on the main subject, rich contrast, and restrained colour.
Silhouette	Create a silhouette of [subject] against [background], strong backlight, clear outline, minimal internal detail, and a striking graphic composition.
Fog or mist	Create a moody image of [subject] in light mist, soft atmospheric depth, muted colours, and clear subject separation.
Reflections	Create an image of [subject] with realistic reflections on [glass/water/polished surface], controlled highlights, and no distorted duplicate objects.

Illustration and Design Effects

Effect	Command
Modern editorial illustration	Create a modern editorial illustration of [concept], clean shapes, balanced composition, warm human details, and a professional technology-publication feel.
Flat vector-style	Create a flat vector-style illustration of [concept], simple geometry, clear hierarchy, limited colour palette, and no tiny text.
Isometric	Create an isometric illustration of [system/process/place], clean 3D-like geometry, labelled zones without tiny text, and a polished SaaS visual style.
Watercolour	Create a watercolour illustration of [subject], soft washes, visible paper texture, gentle edges, and natural colour variation.
Ink drawing	Create a black ink drawing of [subject], expressive linework, controlled detail, white background, and a clean editorial composition.
Clay render	Create a clay-style 3D render of [subject], rounded forms, matte texture, soft shadows, and a friendly product explainer look.
Paper cutout	Create a layered paper cutout illustration of [scene], visible paper edges, soft shadows between layers, and bright but balanced colours.
Comic panel	Create a clean comic-panel image of [scene], bold linework, expressive composition, controlled halftone texture, and no speech bubbles unless requested.
Pixel art	Create pixel art of [subject], limited palette, crisp square pixels, readable silhouette, and a [16-bit / 8-bit] game-inspired feel.
Blueprint diagram	Create a blueprint-style diagram of [object/system], white linework on deep blue background, clean structure, and only short readable labels.

Colour, Texture, and Finish Effects

Effect	Command
Duotone	Create a duotone image of [subject] using [colour 1] and [colour 2], strong contrast, clean shapes, and poster-like clarity.
Monochrome	Create a monochrome image of [subject], rich tonal range, clean contrast, and an elegant black-and-white finish.
Pastel	Create a soft pastel image of [subject], gentle colours, low contrast, airy composition, and a calm approachable mood.
Vibrant campaign	Create a vibrant campaign image of [subject], bold complementary colours, clear focal point, energetic mood, and clean negative space for copy.
Film grain	Create a realistic photo of [subject] with subtle film grain, natural colour response, soft highlights, and an understated analogue feel.
Glossy 3D	Create a glossy 3D render of [object], polished material, realistic reflections, clean studio lighting, and a premium technology aesthetic.
Matte finish	Create a matte-finish product image of [object], soft texture, controlled shadows, subdued highlights, and a calm premium mood.
Vintage print	Create a vintage print-inspired image of [subject], aged paper texture, muted inks, slight registration imperfections, and original generic design language.

Marketing, Web, and Social Media Effects

Use case	Command
Blog hero image	Create a 16:9 blog hero image for an article about [topic]. Show [visual metaphor], professional business context, clear focal point, and space for headline overlay.
Website banner	Create a wide website banner for [service/product], realistic modern business environment, strong subject on one side, clean negative space on the other, and no embedded text.
Social post	Create a 1:1 social media image about [message], bold simple visual metaphor, high contrast, readable composition on mobile, and no small text.
Vertical story	Create a 9:16 vertical story image for [campaign], strong central subject, clean top and bottom safe zones, vibrant lighting, and no UI elements.
YouTube thumbnail concept	Create a YouTube thumbnail concept about [topic], expressive central subject, high contrast, simple background, and space for 2-4 words of large text.
Infographic	Create a clean infographic explaining [process], with 4-6 steps, sharp short labels, simple icons, strong hierarchy, and a 16:9 layout.
Product lifestyle scene	Create a lifestyle image of [product] being used by [audience] in [setting], natural behaviour, realistic lighting, and clear product visibility.
Ad concept	Create a digital ad concept for [offer], premium but practical visual style, clear emotional benefit, space for headline and CTA, and no fake logos.

Image Editing Commands for Existing Photos

ChatGPT can edit a generated image or an uploaded image by selection or by direct instruction. The important habit is to say exactly where the edit applies and what should remain unchanged.

Edit	Command
Remove an object	Remove [object/person/text] from this image. Fill the background naturally and keep the lighting, perspective, and surrounding objects unchanged.
Replace a background	Replace the background with [new background]. Keep the subject's pose, proportions, clothing, lighting direction, and edge detail realistic.
Change colour	Change only the [object] colour to [colour]. Keep the material texture, highlights, shadows, shape, and all other objects unchanged.
Add an object	Add [object] to the [location in image]. Match the existing perspective, lighting, shadow direction, and scale.
Clean up product image	Clean up this product photo by removing dust, distracting marks, and background clutter. Keep the product shape, colour, labels, and proportions unchanged.
Make transparent cutout	Turn the subject into a transparent-background cutout with crisp edges and a natural soft shadow. Do not change the subject.
Extend image	Extend this image to [aspect ratio] by naturally continuing the background and composition. Keep the original subject unchanged.
Restyle image	Restyle this image as [effect], but preserve the subject identity, pose, composition, and important details.
Improve text layout	Recreate this graphic with cleaner typography. Keep the same message, use larger readable text, improve spacing, and avoid adding new words.
Brand-safe generic version	Create a generic, ownable version of this visual. Keep the business concept but remove any recognisable brand, logo, trademark, or specific product styling.

Commands for Text Inside Images

Text in generated images is improving, but it still benefits from very precise prompting. Keep the copy short, put the exact words in quotes, and tell ChatGPT where the text should go.

Goal	Command
Headline poster	Create a poster with the headline "[SHORT HEADLINE]" at the top in large bold sans-serif text. Add no other text. Use [visual style] and [aspect ratio].
Labelled diagram	Create a simple labelled diagram of [process] with exactly [number] labels: "[label 1]", "[label 2]", "[label 3]". Keep labels large and sharp.
Spelling control	Add the brand name "[BRAND]" exactly as written. Spell it letter by letter as [B-R-A-N-D]. Do not add extra letters or words.
No text	Create the image with no text, no captions, no signs, no logos, and no watermark-like marks.

Advanced Commands for More Control

Use multiple images

When using references, give each image a job. For example: Image 1 is the product. Image 2 is the lighting reference. Image 3 is the colour mood. Create a new 16:9 campaign image using the product from Image 1, the lighting from Image 2, and the palette from Image 3.

Ask for staged refinements

Instead of asking for everything at once, work in steps: first generate the core composition, then refine lighting, then fix text, then change the aspect ratio. OpenAI's image guidance recommends small targeted revisions because they are easier to follow and reduce drift between versions.

Preserve identity and product details

For people, products, packaging, uniforms, and places, use preservation language: Preserve the face, pose, product shape, logo placement, colour, material, and camera angle. Change only the background to [new background].

Use negative instructions carefully

Negative instructions are useful when they are specific: no fake logos, no extra text, no distorted hands, no unreadable labels, no brand imitation, no extra objects. Avoid long lists of vague negatives, because they can make the prompt harder to follow.

Safety, Copyright, and Brand Cautions

For business use, do not ask ChatGPT to imitate a living artist, copy a specific brand's visual identity, create misleading likenesses, or fabricate evidence-like images. OpenAI's Academy guidance recommends generic or ownable versions of a design rather than imitating a specific brand, product, or artwork. OpenAI's Usage Policies also place boundaries around privacy, likeness, deception, and harmful uses.

Practical rule: use ChatGPT images for original concepts, drafts, internal exploration, approved campaign art, and controlled edits. For final public assets, review the image for accuracy, licensing risk, accessibility, brand consistency, and whether any in-image text is readable.

Sources Checked

• OpenAI Academy: Creating images with ChatGPT
• OpenAI Help Center: Images in ChatGPT
• OpenAI API documentation: Image generation
• OpenAI Usage Policies