AI product photo generator for marketplace listings
To make marketplace product photos with AI, upload your product shot to the Photo section as a reference and describe the result: a plain white background, a kitchen or bathroom scene, a person holding the item, or an infographic with captions. Ask the model to keep the packaging and text as they are, then check the fine print yourself. One run gives up to four variants, and the best one can become a short video for the listing.
Updated:
Models that edit product photos
| Version | Developer | Modes | Resolution | Aspect ratios | References | Prompt | Features | Price from |
|---|---|---|---|---|---|---|---|---|
| Kling O3 Image | Kling | text to image, edit from photos | 1K, 2K, 4K | 1:1, 9:16, 16:9, 21:9, 2:3, 3:2, 3:4, 4:3 | up to 10 images | up to 4,000 characters | from 2 tokens | |
| Kling O1 Image | Kling | edit from photos | 1K, 2K | 1:1, 9:16, 16:9, 21:9, 2:3, 3:2, 3:4, 4:3 | image required, up to 4 | up to 4,000 characters | from 2 tokens | |
| Nano Banana 2 | text to image, edit from photos | 1K, 2K, 4K | 1:1, 9:16, 16:9, 21:9, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4 | up to 14 images | up to 4,000 characters | from 6 tokens | ||
| Nano Banana Pro | text to image, edit from photos | 1K, 2K, 4K | 1:1, 9:16, 16:9, 21:9, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4 | up to 4 images | up to 4,000 characters | from 12 tokens | ||
| Nano Banana | text to image, edit from photos | 1:1, 9:16, 16:9, 21:9, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4 | up to 4 images | up to 4,000 characters | from 4 tokens | |||
| GPT Image 2.5 Flare | OpenAI | text to image, edit from photos | 1K, 2K, 4K, custom size | 1:1, 9:16, 16:9, 3:4, 4:3, 2:3, 3:2, 4:5, 5:4, 21:9 | up to 16 images | up to 32,000 characters | 5 quality levels | from 2 tokens |
| GPT Image 2.5 Sunburst | OpenAI | text to image, edit from photos | 1K, 2K, 4K, custom size | 1:1, 9:16, 16:9, 3:4, 4:3, 2:3, 3:2, 4:5, 5:4, 21:9 | up to 16 images | up to 32,000 characters | 5 quality levels | from 2 tokens |
| GPT Image 2 | OpenAI | text to image, edit from photos | 1K, 2K, 4K, custom size | 1:1, 9:16, 16:9, 3:4, 4:3, 2:3, 3:2, 4:5, 5:4, 21:9 | up to 16 images | up to 4,000 characters | 3 quality levels | from 2 tokens |
| GPT Image 1.5 | OpenAI | text to image, edit from photos | 1:1, 2:3, 3:2 | up to 1 images | up to 4,000 characters | good at rendering text in images | from 2 tokens | |
| GPT-4o Image | OpenAI | text to image, edit from photos | 1:1, 2:3, 3:2 | up to 2 images | up to 4,000 characters | from 2 tokens | ||
| Seedream 5.0 Pro | ByteDance | text to image, edit from photos | 2K | 1:1, 9:16, 16:9, 3:4, 4:3, 3:2, 2:3, 21:9 | up to 10 images | up to 4,000 characters | from 8 tokens | |
| Seedream 5.0 Lite | ByteDance | text to image, edit from photos | 2K, custom size | 1:1, 9:16, 16:9, 3:4, 4:3, 3:2, 2:3, 21:9 | up to 4 images | up to 4,000 characters | from 3 tokens | |
| Seedream 4.5 | ByteDance | text to image, edit from photos | 2K, 4K, custom size | 1:1, 9:16, 16:9, 3:4, 4:3, 3:2, 2:3, 21:9 | up to 8 images | up to 4,000 characters | from 4 tokens | |
| Seedream 4 | ByteDance | text to image, edit from photos | 2K, 4K | 1:1, 9:16, 16:9, 3:4, 4:3, 3:2, 2:3, 21:9 | up to 8 images | up to 4,000 characters | from 4 tokens | |
| FLUX.2 | Black Forest Labs | text to image, edit from photos | 1K, 2K | 1:1, 9:16, 16:9, 3:4, 4:3, 3:2, 2:3 | up to 8 images | up to 4,000 characters | from 3 tokens | |
| Flux Kontext | Black Forest Labs | text to image, edit from photos | 1:1, 9:16, 16:9, 3:4, 4:3, 21:9, 9:21 | up to 1 images | up to 4,000 characters | from 4 tokens | ||
| Wan 2.7 Image | Alibaba | text to image, edit from photos | 1:1, 9:16, 16:9, 3:4, 4:3 | up to 4 images | up to 4,000 characters | from 2 tokens |
How to make a listing image with AI
Shoot the product
A phone photo in even light is enough: the whole item in frame, the label readable, no glare on the packaging. A white background helps but is not required.
Upload it and describe the shot
In the Photo section, add the picture to the references and say what background, scene or captions you need. In a separate sentence, ask to keep the shape, color, logo and packaging text as they are.
Pick the format and quantity
Choose the aspect ratio your marketplace expects and set Quantity to anywhere from 1 to 4. The token price is visible before you run.
How to keep the product accurate
The model redraws the whole frame, and your photo is its main guide: the sharper it is, the less gets made up. If several sides of the item matter, shoot each one and add every shot to the references. GPT Image 2 and 2.5 take the most at once (up to 16), followed by Nano Banana 2 (up to 14).
In the prompt, list what must not change: shape, color, logo, label text, the number of pieces in a set. GPT Image 2.5 Sunburst and Nano Banana Pro suit careful edits from a photo, and Kling O1 Image will not run without one at all: it only edits.
Models sometimes garble small print on packaging, so compare the result with the original before it goes to a marketplace. If only the text went wrong, send the image back with "Use as a reference for photo" and ask to fix just that part.
Background, scene and a person in the shot
For the main image, a clean white or light plain background is the usual choice. Say so, add a soft shadow under the item and rule out any other objects. Extra images can show the product in use: on a kitchen counter, a bathroom shelf or a desk, or in a studio with a colored backdrop. Describe the surface, the light and what sits nearby: that helps the model keep the item at its real scale.
To show a person holding the product or wearing the clothes, add their photo to the references or pick a character from the Characters section: the character photos go in next to your product shot. Choose a model with room for plenty of references, such as GPT Image 2 or Nano Banana 2, so the person and the product both fit. Check hands and fingers closely, as models often get them wrong.
Infographics: writing text on the image
Put each caption in quotes and write it exactly as it should appear, letters and numbers included. Keep one idea per callout: a few short points read better than a paragraph. Say where the text goes, what color it is and what it sits on, so it does not cover the product.
GPT Image 1.5 and GPT-4o Image handle crisp lettering well, but they offer only three formats: 1:1, 2:3 and 3:2. Write only what is true about the product, and check every letter and number before publishing.
A series in one style, plus other angles
Set Quantity anywhere from 1 to 4 and one run returns several takes on the same prompt, enough for a main image and a few extra ones. For the rest of the line, keep one prompt template and change only the product name and color. Add the best finished image to the next run as a style sample: it helps keep the light, background and layout consistent across the series.
Ask for angles directly: top view, side view, three-quarter view, a close-up of a detail. Any side that is missing from your photos will be invented by the model, so shoot the back, the fasteners and the seams yourself.
What to check against marketplace rules
- Main image: the product itself, large and whole. Amazon asks for a pure white background here, while Ozon and Wildberries have their own rules, and they change, so check them in your seller account.
- Honest text: no "No. 1" or "bestseller" badges, no discounts that do not exist, no logos or certification marks that are not yours. A marketplace can reject such images.
- A match with the real item: the color, size and contents shown should be what the buyer receives. Props from the scene that are not in the box are better kept off the main image.
- Pixel size and file weight: an exact size in pixels can be typed in on Seedream 4.5, Seedream 5.0 Lite, GPT Image 2 and both GPT Image 2.5 versions. If a marketplace limits file weight, a finished image has a Compress button.
A video for the listing
A finished image can become the first frame of a clip: press "To video" on it and it opens in the Video section. Then describe the motion: a slow turn of the product, a camera push-in, a hand picking the box up off the table. The product ad video page covers full ad clips in more detail.
More for sellers
Where to find it in ClickBait
Marketplace product photos lives in ClickBait under the Photo section. Sign up, pick a version from the model list and describe what you want to get.
The token price is shown on the run button before you start. If a generation fails, the tokens go back to your balance automatically.
New accounts created with Google or email get a welcome bonus of 10 tokens. These are promo tokens and expire after 14 days.
Models for this task
Kling
video 3-15 s; matches the source video, up to 4K
images up to 4K
up to 10 references
from 2 tokens
Nano Banana
images up to 4K
up to 14 references
from 4 tokens
GPT Image
images up to 4K
up to 16 references
from 2 tokens
Seedream
images up to 4K
up to 10 references
from 3 tokens
FLUX
images up to 2K
up to 8 references
from 3 tokens
WAN
video 2-30 s, up to 1080p
up to 4 references
from 2 tokens
Other tools
AI video generator
12 models
Kling, Seedance, Veo, WAN, MiniMax H3
from 8 tokens
Image to video
13 models
Kling, Kling Motion Control, Seedance, Veo, WAN, MiniMax H3
from 8 tokens
AI image generator
18 models
Kling, Nano Banana, GPT Image, Seedream, FLUX, WAN, Z-Image
from 2 tokens
AI voiceover
5 models
ElevenLabs, Minimax Speech-02 HD, Seed Audio 1.0
from 2 tokens
Lip sync
5 models
Lipsync 2, Lipsync 2 Pro, Lipsync 3, InfiniteTalk, LatentSync
from 4 tokens
AI music generator
9 models
Suno
from 5 tokens
Video upscaler
1 model
Upscale
from 2 tokens
AI UGC video
12 models
Kling, Seedance, Veo, WAN, MiniMax H3
from 8 tokens
Product ad video
12 models
Kling, Seedance, Veo, WAN, MiniMax H3
from 8 tokens
Creatives for affiliate marketing
30 models
Kling, Seedance, Veo, Nano Banana, GPT Image, Seedream, FLUX, WAN, MiniMax H3, Z-Image
from 2 tokens
Videos for TikTok and Reels
12 models
Kling, Seedance, Veo, WAN, MiniMax H3
from 8 tokens
Ad posters and banners
17 models
Kling, Nano Banana, GPT Image, Seedream, FLUX, WAN, Z-Image
from 2 tokens
Frequently asked questions
How do I make marketplace product photos with AI?
Open the Photo section, add your product shot to the references and describe the background, scene or captions. Then pick a model that edits from photos, an aspect ratio and the number of variants, and run it.
How do I change the background of a product photo?
Describe the new background in words, for example: "the same product on a clean white background, soft shadow below, no other objects". The AI does not cut the item out, it redraws the whole frame, so compare the shape and color with the original.
Will the AI change the packaging or the label?
It can, especially small print. The risk drops when the photo is sharp, the prompt asks to leave the packaging alone and the references include shots from different sides. A wrong caption can be fixed in a separate run by sending the image back to the references.
How do I make product infographics with AI?
Type the caption text word for word in quotes and say where each caption goes and in what color. GPT Image 1.5 and GPT-4o Image are good picks for lettering. Check every letter and number before you upload.
Can I use these images on Amazon, Ozon or Wildberries?
Yes, as long as they follow the marketplace rules. Each one sets its own requirements for the main image, background and text, and they change from time to time, so check them in your seller account before uploading.
How do I get several variants and a whole line in one style?
Set Quantity to anywhere from 1 to 4 variants per run. For the next products, repeat the same prompt and add the best finished image to the references as a style sample.
Can I make a video for the listing?
Yes. Press "To video" on a finished image and it becomes the first frame of a clip in the Video section. Then describe the motion, for example a slow turn of the product.