Skip to main content

How to generate and edit images in Sintra AI?

C
Written by Cassandra

Good visuals sell. Product shots, social graphics, banners, and ads all need images, and most small businesses don't have a designer on call. Sintra AI lets you create new images and edit the ones you already have, right in the chat, with your brand colors and logo when you want them.

This article covers how to create an image, how to edit an existing photo, how to keep your branding consistent, what images cost in credits, and answers to the questions customers ask most.


What Sintra AI can do with images

  1. Create new images from a description. Describe what you want, pick a shape (square, portrait, landscape, or wide), and an AI Employee creates the image for you.

  2. Edit images you already have. Upload a photo and describe the change: a new background, brighter colors, something added or removed, or your logo in a corner.

  3. Use your Brand Kit. AI Employees can use the logo and brand colors saved in your Brand Kit in Brain AI, so you don't need to re-upload them for every image.

  4. Work from your real photos. Give Sintra AI reference photos of your product, your team, or your premises, and it creates realistic images based on them, including clean product shots on a white background.

  5. Combine photos. Several photos can be merged into one image, for example your product placed in a new setting.

Who does the work? Any AI Employee can create and edit images, and so can Team chat. You don't need to pick a specific AI Employee.

Where? In the web app and in the Sintra AI mobile app for iPhone, iPad, and Android, including uploading and editing your own photos.


Example use cases

  • A clean product shot from a phone photo. Snap your product on the kitchen table, upload the photo, and ask for an e-commerce-ready image on a white background, in your brand style.

  • A branded social graphic for a promotion. Ask for a square image announcing your weekend offer, in your brand colors, with your logo in the corner and space for a caption.

  • A banner for your website or newsletter. Ask for a wide (16:9) image that matches your website's look, then ask for a second version with different text.

  • A new background for an existing photo. Upload a photo of your team at the office and ask to replace the busy background with a plain, bright one. Photos of real people can be rejected by the content policy, so this won't work with every team photo.

  • A set of images for a campaign. Ask for 5 images with the same color palette and illustration style, one per product, describing the shared style in each request.

  • A product placed in a lifestyle scene. Upload 2 or 3 photos of your product from different angles and ask to show it on a café table in morning light.

  • A quick fix before posting. Ask to brighten a dark photo, crop it to portrait, and add your logo, all in one request.


How image requests work

You describe, the AI Employee creates. Write what you want in plain words, as you would brief a designer: the subject, the style (photo, illustration, or flat design), the colors, the background, and the mood. The more specific you are, the closer the first result.

Each image is a fresh start. A new image doesn't automatically remember the one before it. To keep a series consistent, repeat the shared style and colors in each request, or say what you liked about the last image.

Edits are requests too. To change an image, reply in the same chat with what should be different, such as "make the background white", "move the logo to the top right", or "make it brighter". You can also open the image and paint over the area you want to change, on web and mobile. Every new version is a new image and uses credits again, so a clear first brief saves credits.

Reference photos guide the result. Photos you upload guide the style and content of the image, and real products look most accurate when you give several clear photos from different angles. Always check small details such as text, labels, and logos before you use an image, and ask for a correction if something is off.


Step by step: creating your first image

Step 1: Open a chat

Open a chat with any AI Employee, or use Team chat. Make sure you're in the right workspace, because AI Employees use that workspace's Brand Kit and Brain AI.

Step 2: Describe the image

Write what you want, including the style, colors, setting, and anything that must appear. Say which shape you need:

  • Square (1:1) for most social posts

  • Portrait (9:16) for stories and phone screens

  • Landscape (4:3) for websites

  • Wide (16:9) for banners

Instagram's 4:5 feed shape isn't available, so the closest shape is used, and publishing can crop the edges, including a logo. For example:

"Create a square image for Instagram: a flat-lay of our iced latte on a light wooden table, soft morning light, in our brand colors, with space at the bottom for a caption."

Tip: Mention composition if it matters, for example "centered", "lots of empty space on the left", or "white background".

Step 3: Review and adjust

The image appears in the chat once it's ready, which can take up to a few minutes. If something isn't right, say exactly what to change. You can also ask for a new version of the same brief. For example:

"Keep everything the same but make the background white and the cup bigger."

Step 4: Save or download

Download the image, or save it to Brain AI so your AI Employees can reuse it later. See "Save, reuse, and download your images" below.


Add your logo and brand colors

Fill in your Brand Kit in Brain AI once (logo, colors, brand voice, description, and unique value proposition).

  • Brand colors: AI Employees use your brand colors in images where they fit.

  • Brand voice: Your brand voice is used for writing, not for images.

  • Logo: AI Employees don't add your logo to every image, so ask for it when you want it. You can also upload your logo file directly in the chat if you want a different version of it, or say where it should go. For example:

"Add our logo from the Brand Kit to the top right corner of this image, small, and use our brand colors for the text."


How your logo appears in images

  • Redrawn by default: In a new image, your logo is redrawn as part of the image, so it can look slightly different from your file.

  • Exact logo file: To place your exact logo file instead, ask for your exact logo, say the drawn logo isn't right, or ask to put your logo on a photo you uploaded.

  • Best file type: For exact placement, a PNG with a transparent background gives the cleanest result. JPEG, WebP, and GIF also work, but SVG doesn't. A logo without a transparent background sits in a box, and a low-resolution file looks soft.

  • Moving or resizing: Moving or resizing an exact logo doesn't create a new image, but changing the scene does.

  • Missing logo: If your logo was picked up from your website during Brand Kit setup and is missing, it may have been an SVG or smaller than 64px, so upload a larger PNG.


Edit an image you already have

Upload a photo (PNG, JPEG, GIF, or WebP) from your device into the chat and describe the change you want. For example:

"Here's a photo of our new shop front. Remove the parked car, brighten the image, and crop it to a square."

You can ask for:

  • A new or plain background

  • Brighter or warmer colors

  • A different crop or shape

  • Something added (your logo, a border, or text)

  • Something removed

Cut-outs and upscaling: You can also ask to remove the background completely for a cut-out, or to upscale a small or blurry image by 2× to 6×. Both give you a PNG file.


Turn phone snaps into product photos

Upload photos of your product from several angles. Ask for clean, e-commerce-ready images on a white background, or in a setting that fits your brand. For example:

"Here are 4 photos of our ceramic travel mug. Create a clean product photo on a white background, front view, studio lighting, true to the real colors and shape."

  • Lighting and background: Lighting and background in your photos can still affect the shadows and the product, even when you ask for a new background, so clear, evenly lit photos give the closest result.

  • Number of photos: Give several photos if you want the product to look as accurate as possible, up to 10 per request.


Combine several photos into one image

Upload 2 to 10 photos and describe how they should come together, for example your product in a scene you photographed, or 2 products side by side. For example:

"Place the mug from the first photo on the café table in the second photo, matching the lighting."


Create a series of images in one style

Each image is created separately, so describe the shared style in every request: the color palette, the illustration or photo style, and the layout. For example:

"Create a landscape image for our 'Autumn menu' campaign: flat illustration style, warm orange and cream palette, one dish in the center, no text. This is the first of 5 in the same style."

Using a look you've found elsewhere: Drop a few screenshots of visuals you like into the chat, discuss the direction, and ask the AI Employee to save a description of it to Brain AI so you can refer to it later. New images don't follow it on their own, so repeat the style in each request when the series has to match. Your written brand voice stays as it is in Brain AI.


Add text to an image

Ask for the exact words to appear on the image, in quotes, and say where they should go. For example:

"Add the text 'Open Sundays from 10am' in white at the bottom of this image."

The text is drawn as part of the image and there's no brand font to apply, so check the spelling in the result before you use it, and check that names and brand words are spelled correctly in Brain AI too.


Change the image in a scheduled post

AI Employees can create, schedule, and publish posts to your connected social accounts, with your approval. Before a post goes out, you can change its image in 2 ways.

  1. In the post itself: Use Regenerate for a new AI image, AI Edit for a specific change such as "make it brighter" or "add our logo", or replace the file with your own photo from your device or from Brain AI. These don't use credits.

  2. In the chat: Ask the AI Employee to swap the image for a file from Brain AI or an image link, or to create a new image for the post. Swapping the image doesn't create a new image, but asking for a new or edited image uses credits like any other image. For example:

"Replace the image in Thursday's post with the 'Autumn menu – pumpkin soup' photo from Brain AI and keep the caption as it is."

The post stays pending until you approve it, so check the preview shows the new image first.


Save, reuse, and download your images

  • Download: Use the download option under the image in the chat to save it to your device.

  • Save to Brain AI: Use Save to Brain under the image, on web or mobile, to save it to Brain AI for the active workspace, so every AI Employee there can use it later, for posts, your website, or new images.

  • Reuse by title: Give saved images clear titles ("Autumn menu – pumpkin soup – top view"), then describe the image in your request. The title doesn't have to match word for word. For example:

"Create a LinkedIn post using the 'Team photo – October 2026' image from Brain AI."

Images with your exact logo: An image with your exact logo placed on it appears as a file in the chat instead, without these buttons. On web, use the file card to download it, or open the file for Download and Save to Brain. On mobile, open the file to download it or save it to Brain AI. You can also ask in the chat to save it to Brain AI.


Good to know

Sintra AI creates images from your description and your reference photos. It can't guarantee a pixel-perfect copy of a real product, so always check small details such as labels, logos, text, and fine shapes, and ask for a correction if something's wrong. Every new version uses credits, so put the important details in your first request.

AI Employees can't take product photos from your online store or a product page link, but they can use a direct link to an image file that you paste in the chat. Otherwise, upload the product photos you want to use, or save them in Brain AI and refer to them by title.


Credits

Creating an image in the chat uses about 0.5 to 1.8 credits. Credit costs are approximate. They're the same for every AI Employee in Sintra AI and the marketplace, but can vary with the task and the information you give in the chat.

  • Every version counts: Each new image, including a regenerated or edited version, uses credits again. A clear brief keeps the total low.

  • No refunds for images you don't like: Credits are used when an image is created, whether or not you use it. If an image fails or is rejected by the content policy, it isn't charged as an image, but the chat message still uses credits.

For the full breakdown, see Workspace credits.


Tips for better images

  • Describe the image the way you'd brief a designer: subject, style, colors, background, mood, and shape.

  • Fill in your Brand Kit first, so your brand colors and logo are ready to use.

  • Upload real photos of your product whenever accuracy matters, from several angles.

  • Put the exact words in quotes when you want text on the image, and check the spelling in the result.

  • Ask for one change at a time when fine-tuning, so you can see what each change did.

  • Keep the original photo; edits create a new image and don't overwrite your file.

  • If you don't want new images created for your posts, attach your own photos or use ones saved in Brain AI.


FAQ

How many credits does it use to create or edit an image?

About 0.5 to 1.8 credits per image in the chat. Credit costs are approximate and can vary with the task and the information you give in the chat.

Will it show my real product accurately, for example a car, a watch, or a staircase, instead of inventing details?

It gets closest when you upload clear photos of the real product from several angles and describe what must stay the same. Reference photos guide the result, so check small details such as badges, text, and proportions before you use the image, and ask for a correction if needed.

Can it pull product images from my store or a product link?

No. AI Employees can't take photos from your store or a product page, but they can use a direct link to an image file that you paste in the chat. Otherwise, upload the product photos you want to use, or save them in Brain AI with descriptive titles and refer to them by title.

Can I give it a photo and have it enhance or change the background?

Yes. Upload the photo and describe the background you want: plain white, blurred, a new setting, or brighter and cleaner. You can also ask to remove the background completely for a PNG cut-out, or to upscale the image by 2× to 6×.

Which AI Employee should I ask for images?

Any of them. Image creation works with every AI Employee and in Team chat. Pick the one you're already working with.

Can I create a series of images with the same theme?

Yes. Describe the shared style and colors in each request, because a new image doesn't automatically remember the previous one.

Why isn't my logo appearing, or why does it look different?

A logo isn't added to every image, so ask for it and say exactly where it should go. Make sure your logo is saved in the Brand Kit in Brain AI, or upload the logo file in the chat. In a new image, your logo is redrawn, so it can look slightly different from your file. To place your exact file, ask for your exact logo or say the drawn one isn't right. A PNG with a transparent background works best. JPEG, WebP, and GIF also work, but SVG doesn't. If your logo was picked up from your website and is missing, it may have been an SVG or smaller than 64px, so upload a larger PNG.

Can I add text to an image?

Yes. Put the exact words in quotes and say where they should appear. The text is drawn as part of the image and there's no brand font to apply, so check the spelling in the result.

Which image shapes can I choose?

Square (1:1), portrait (9:16), landscape (4:3), or wide (16:9). Say which one you need, or name the platform ("for an Instagram story"). Instagram's 4:5 feed shape isn't available, so the closest shape is used, and publishing can crop the edges, including a logo. When you edit a photo, it keeps its original shape unless you ask for another one.

What resolution and file format do I get?

Images in the chat are PNG files at 1K resolution, and images made with Regenerate or AI Edit in a scheduled post are also 1K.

Can I upload iPhone photos (HEIC) to edit?

Convert them to JPG or PNG first, for example by sharing or exporting the photo from your Photos app as JPG. AI Employees can only see PNG, JPEG, GIF, and WebP images directly. You can save a HEIC file in Brain AI, but convert it if you want to edit it.

Can I use my own photos instead of AI images in posts?

Yes. Attach the photo in the chat when you ask for the post. That photo is used as the image, and it is not replaced with an AI image. You can also use a photo already saved in Brain AI.

How do I stop an AI Employee from creating images I didn't ask for?

For posts, give it the photo to use. Attach it in the chat, or use one already saved in Brain AI, and the post uses that photo. A post can't be scheduled without an image, so if you don't supply one, an image is created. For anything else, say in your request that you don't want any images created.

Do I pay if I don't like the image?

Yes. Credits are used when an image is created, and new versions use credits again. A detailed first request is the best way to keep costs down.

Can I make and edit images on my phone?

Yes, in the Sintra AI mobile app for iPhone, iPad, and Android. You can create images, upload and edit photos, paint over an area to change it, upscale, remove backgrounds, and download, save, or share the result.

Where do my images go after I save them?

Saved images go to Brain AI for the active workspace, where every AI Employee in that workspace can use them. Downloaded images are saved to your device.

Did this answer your question?