Amazon product image requirements (and how to meet them at scale)
Amazon product image requirements are the rules Amazon sets for the photos on a product detail page. The strictest apply to the main image, the one shoppers see in search results: a photograph of the product alone, on a pure white background, filling at least 85% of the frame, with no text or graphics. Additional images have more freedom, and A+ Content has its own specifications.
For one listing, the rules fit on a page. For a brand with thousands of SKUs on Amazon, Walmart and other retailers, the difficulty is that each retailer’s rulebook is slightly different, packs keep changing, and every image has to be right for every one of them. This guide sets out Amazon’s 2026 requirements, compares them with Walmart’s and GS1’s, and covers how to meet them at scale.
One note on sources. Amazon’s own help pages on image requirements sit behind Seller Central and change without notice. The rules below come from Amazon’s published guides and from guidance posted by Amazon staff on its seller forums. Check Seller Central for your category before a large upload.
Amazon main image requirements
The main image decides how a product appears in search, and it is the one Amazon enforces most strictly:
- Pure white background. RGB 255, 255, 255, with no off-white, gradients or scenes.
- The product fills at least 85% of the frame. Amazon staff guidance in 2026 gives 85% to 100%.
- A photograph of the actual product. No illustrations or placeholder images.
- Only what is sold. No props or accessories that are not included, and no multiple views of the product in one image.
- Nothing added. No text, logos, graphics, watermarks, promotional messages, borders, colour blocks or inset images, and no badges.
- The whole product in frame. The product is not cropped at the edges.
- Packaging where it informs the shopper. Amazon’s guidance is to show the product outside its packaging unless the box, bag or case is a key feature. Packaging is acceptable for products such as consumables and collectibles. For multipacks, the main image should show the product in its packaging and the total quantity being sold, and the title should state the pack count.
Clothing has its own rules. Amazon’s clothing style guide asks for adult clothing on a standing human model, while children’s clothing is shown flat, without a model.
Technical requirements
Guides online quote several different sizes. Amazon’s staff guidance and style guides give the following:
| Specification | Requirement |
|---|---|
| Minimum size | 500 pixels on the longest side |
| Size for zoom | 1,000 pixels or more on the longest side |
| Optimal size | 1,600 pixels or more |
| Maximum size | 10,000 pixels on the longest side |
| Aspect ratio | No fixed ratio; Amazon adds white bars to non-square images |
| File formats | JPEG (preferred), TIFF, PNG and non-animated GIF |
| Colour mode | RGB preferred; CMYK is accepted but can shift colour on screen |
| Number of images | One main image plus up to eight more; Amazon recommends at least six and a video |
| File names | Product identifier, then a variant code, then the extension, for example B00SAMPLE.PT02.jpg |
The product identifier in the file name can be the ASIN, UPC, EAN, JAN or ISBN, without dashes or spaces. MAIN marks the main image and PT01 to PT08 the additional ones. Images under 1,000 pixels don’t support zoom, which shoppers use to read labels and inspect details.
Additional images
Additional images can show the product from other angles, in use, in a setting, with props, at scale and with infographics that explain features or dimensions. Promotional text such as “free shipping” or “sale” is not allowed. Amazon’s guides are less consistent about other text on additional images, so keep any text short, factual and readable on a phone.
Some categories add content rules. Trade press reported that dietary supplement listings had until 31 March 2026 to make their images and copy match the product’s Supplement Facts panel.
What’s new in 2026: tagging AI-generated people
In July 2026 Amazon told sellers to tag images and videos that show photorealistic AI-generated people, in listings and in A+ Content, across its stores worldwide. The tag is the keyword contains-synthetic-performer, added to the file’s XMP dc:subject metadata field before upload with an IPTC-compatible editor. Amazon said it will show customers an indicator where applicable.
The rule follows New York’s law on synthetic performers in advertising, which took effect in June 2026. Real people edited with AI, characters from films, TV, games and other creative works, images without people and people who are not photorealistic are exempt. Beyond this tag, Amazon’s rules on AI imagery come down to accuracy: the image has to show the real product and meet the same requirements as any other image. For the wider rules on AI product images, including Google’s metadata requirement and the EU AI Act, see AI product photography.
A+ Content image specifications
A+ Content is the richer section further down the detail page, available to brand-registered sellers. Amazon’s guidance:
- File size. Images up to 2 MB.
- Modules. Up to five modules for Basic A+ and up to seven for Premium A+, which adds video, hotspots and carousels.
- Text in images. Amazon’s 2025 design guide advises avoiding text in images, because it is hard to read on mobile.
- Claimed effect. Amazon says Basic A+ can increase sales by up to 8% and Premium A+ by up to 20%. These are Amazon’s own figures, based on internal data.
What happens when images don’t comply
A missing or non-compliant main image can push a product down in search results or suppress it from search altogether. A common cause is a main image that isn’t on pure white. Amazon staff guidance is that replacing the image with a compliant one lifts the suppression. Amazon doesn’t publish how its automated checks work, so the safest approach is to check every image before upload rather than after a listing disappears.
Amazon vs Walmart vs GS1 image requirements
Most brands sell through more than one retailer, and their rules overlap without matching:
| Requirement | Amazon | Walmart (US) | GS1 Product Image Standard |
|---|---|---|---|
| Background | Pure white, RGB 255 | Pure white, RGB 255 | White, RGB 255, or transparent |
| Product fill | At least 85% of the frame | Square image; Walmart Canada asks for a 2.5% white border | About 95% of the canvas |
| Size | 1,000 px or more for zoom, 1,600 px optimal | 2,200 × 2,200 px recommended; 1,500 × 1,500 px for zoom | 900 to 2,400 px for web images |
| Aspect ratio | Any | 1:1 | Square |
| File names | ASIN or other product identifier plus variant code | 14-digit GTIN recommended | GTIN plus image type, facing, angle and state codes |
| Number of images | Up to 9; at least 6 recommended | At least 4 recommended | One master per view, others derived |
| Formats | JPEG, TIFF, PNG, GIF | JPEG, PNG, BMP; up to 5 MB | Stored as TIFF, shared as JPEG or PNG |
European retailers vary as well. The Dutch retailer bol accepts images from 500 to 6,000 pixels per side and asks for a white background without shadows and the product out of its packaging, although since July 2025 it no longer enforces most of its image guidelines. Many grocers take images through GS1’s data pools, which carry links to the image files rather than the files themselves. Since February 2023, suppliers to German retailers have had to provide at least one image per product through GDSN, and in May 2023 REWE Group found that 42.2% of the items it received through GDSN still had no image.
The differences are small, but they are enough that one file rarely meets every retailer’s rules. An image framed to GS1’s 95% fills more of the frame than Amazon requires, while Walmart US asks for the item to be cropped as close to the frame as possible. A file named by GTIN needs renaming for an Amazon upload, and every manual export adds a chance of error.
How to meet Amazon’s image requirements at scale
At a few dozen SKUs, a checklist and a careful retoucher are enough. At thousands, the work has to be organized around a few principles:
- Keep one master per product view and pack version. Store high-resolution masters to the GS1 standard and keep track of which pack version each one shows, because the GTIN often stays the same when the artwork changes. See packshots at scale for how pack versions are tracked.
- Treat each retailer’s rules as data. Background, fill, size, aspect ratio, border, naming and format become rules that produce each retailer’s version from the master, rather than instructions someone follows by hand.
- Validate before upload. Check every file against the target retailer’s rules: background values, fill percentage, pixel size, file name and metadata, including the AI tag where needed.
- Regenerate when the pack changes. When artwork, claims or pack sizes change, identify every image that shows the old pack, across every retailer, and replace it.
- Watch for rule changes. Retailer requirements change every year. The AI-generated people tag and GS1’s new AI flag both arrived in 2026, and rules kept as data can be updated in one place.
What it looks like in practice
Grip validates every e-commerce asset against GS1, Amazon and Walmart standards by default, and against Carrefour and other retailers’ requirements, to protect product pages from suppression and delisting. Facing decentralized workflows, rising localization costs and strict brand and compliance demands, Opella turned to Grip to centralize its e-commerce content production while giving local teams room to move fast.
Grip is the enterprise automation platform for publishable marketing content at scale, built by INDG. For retailer requirements, it produces each retailer’s version from the same source. Digital Shelf generates hero shots, alternate views and promo variants from digital twins of the actual product, and automates multipacks at SKU level, so the Amazon main image, the square Walmart image and the GS1 master all show the same, current pack. Art direction and retailer standards are encoded in its Brand Center and applied by default. Grip connects to artwork management, PIM, DAM and syndication systems, so compliant images reach each retailer without a manual export for every file.
Where to start
Export the main images for your top 100 Amazon ASINs and check them against the main image rules above: background value, fill, size, overlays and pack version. Then check the same products on your next two retailers. The number of files that pass on one retailer and fail on another shows how much of the work is in the differences between rulebooks. To see how images fit into the rest of the page, read product detail page content, or talk to the Grip team.
Frequently asked questions
What size should Amazon product images be?
Amazon staff guidance gives 1,000 pixels or more on the longest side as the size that enables zoom, 1,600 pixels or more as optimal, 500 pixels as the minimum and 10,000 pixels as the maximum. JPEG is preferred, and TIFF, PNG and non-animated GIF are also accepted.
How many images can you add to an Amazon listing?
A listing can have one main image and up to eight additional images, nine in total. Amazon recommends at least six images and a video.
Can an Amazon main image show the packaging?
Amazon’s guidance is to show the product outside its packaging unless the box, bag or case is a key feature. Packaging is acceptable for products such as consumables and collectibles, and multipacks should show the full quantity being sold.
Do you have to label AI-generated images on Amazon?
Since July 2026 Amazon requires sellers to tag images and videos in listings and A+ Content that show photorealistic AI-generated people, by adding the keyword contains-synthetic-performer to the file’s XMP metadata before upload. Real people edited with AI, characters from films, TV, games and other creative works, and images without people are exempt.
How do Amazon’s image requirements compare with Walmart and GS1?
All three call for a white background, but the details differ. Amazon asks for the product to fill at least 85% of the frame and has no fixed aspect ratio; Walmart recommends square 2,200 by 2,200 pixel images; GS1’s standard has the product fill about 95% of the canvas and names files by GTIN, while Amazon names them by ASIN or another product identifier.
References
- Amazon Seller Central, product image requirements (requires sign-in).
- Amazon, “Product photos,” Sell on Amazon blog, 4 December 2024.
- Amazon Selling Partner API, Catalog Items API reference (image variants MAIN and PT01 to PT08).
- Amazon Seller Forums, “Product Image Requirements and Best Practices,” Amazon staff post.
- Amazon Seller Forums, “Product Photography Standards: What’s New in 2026,” Amazon staff post.
- Amazon Seller Forums, “Product Image Aspect Ratio,” Amazon staff post.
- Amazon Seller Central Europe forums, “Listings Lounge: Product Image Requirements,” Amazon staff post.
- Amazon Seller Forums, “Quick Tip: Product Images,” Amazon staff post.
- Amazon Seller Forums, multipack image requirements, Amazon staff post.
- Amazon Seller Forums, suppressed items thread, Amazon staff reply.
- Amazon, Selling Partner Style Guide: Clothing & Accessories, March 2022.
- Amazon Seller Forums, “You must now tag AI-generated people in listings and A+ content,” July 2026.
- CNBC, on Amazon’s labels for AI-generated people in product images, 23 July 2026.
- SupplySide Supplement Journal, on Amazon and dietary supplement compliance, 31 August 2026.
- Amazon, “How to create A+ Content: Design tips, guidelines, and examples,” 24 September 2025.
- Amazon, A+ Content.
- Walmart Marketplace, “Product detail page: Image guidelines & requirements,” updated 12 May 2026.
- Walmart Marketplace Canada, “Product Content Page: Image Guidelines & Requirements,” updated 25 February 2026.
- GS1, Product Image Standard, Release 5.1, September 2026.
- bol, “These rules apply to your images.”
- GS1 Netherlands, guidelines for digital files and product images, March 2025.
- GS1 Australia, National Product Catalogue data dictionary, uniform resource identifier attribute.
- Retail Optimiser, “REWE Group requests standardised product images via GDSN,” 15 September 2023.
About Grip
As INDG’s software branch, Grip is a visual content configuration engine powered by NVIDIA Omniverse that makes it possible for large enterprises to use AI at scale. It breaks existing content down into configurable modules, allowing brands to swap out any element, including products, talent, accessories, and branding assets, with complete control and accuracy. Grip integrates with existing workflows to automate product swaps and generate endless, hero-quality content variation, without disrupting established content production processes.
