Quick Answer
Amazon accepts product images from 500 to 10,000 pixels on the longest side and prefers files above 1,000 pixels for zoom; Amazon's 2026 seller guidance calls 1,600 pixels or more the optimal target. The main image must accurately show the actual product on pure white (RGB 255, 255, 255), usually filling at least 85% of the frame, without added text, graphics, watermarks, or confusing props. The safest production plan treats Amazon photography as two systems: a verified real-product main image for truth and compliance, then professional or AI-assisted secondary images for explanation and scale, with human review closing both.
What are Amazon's product image requirements in 2026?
Amazon's current baseline separates file acceptance from main-image content rules. The official Product Image Requirements summary says all images must be 500 to 10,000 pixels on the longest side, clear, and supplied as JPEG, TIFF, PNG, or non-animated GIF. JPEG is preferred.
“We prefer images to be larger than 1,000 pixels on the longest side to allow for zoom.”
That guidance comes from Amazon's Seller Forums team. A separate 2026 Amazon seller guidance post calls 1,600 pixels or more optimal, so 1,600 pixels is the sensible production floor when you want a zoom-capable file rather than the smallest accepted upload.
| Image property | Amazon baseline | Practical production target |
|---|---|---|
| Longest side | 500 to 10,000 pixels | At least 1,600 pixels |
| File type | JPEG, TIFF, PNG, or non-animated GIF | JPEG unless transparency or another workflow requires PNG |
| Image quality | Clear, without pixelation or jagged edges | Inspect at full size before upload |
| Main background | Pure white, RGB 255, 255, 255 | Sample the exported corners, not only the editor canvas |
| Main product coverage | At least 85% of the frame in the general guidance | Fill the frame without cropping the product |
What must the Amazon main product image contain?
The Amazon main image must show the actual item for sale clearly and without additions that could confuse a buyer. Amazon's Product Image Guide and its current seller guidance require or prohibit the following:
- Use a pure white background with RGB values 255, 255, 255.
- Show the entire product and only what the buyer receives.
- Keep the product at 85% or more of the frame under the general rule.
- Remove added text, graphics, borders, watermarks, promotional badges, and inset images.
- Do not use a sketch, placeholder, or mockup in place of the actual product.
- Keep the image in focus, accurately colored, and free of visible distortion.
Amazon may reject, remove, or alter images that fail its requirements, and a non-compliant main image can contribute to search suppression. The final export matters: a white canvas can acquire an off-white edge, compression artifact, or soft shadow during editing.
Do Amazon image rules vary by product category?
Amazon category rules take precedence when they conflict with the general image requirements. The current guidance, reproduced in an Amazon Seller Forums answer quoting the Product Image Requirements, includes examples such as a single shoe facing left at a 45-degree angle, Women's and Men's Clothing photographed on a model, and Kids & Baby Clothing photographed flat and off-model.
Do not treat those examples as a complete category checklist. Open the category style guide linked from Seller Central for the exact product type and marketplace before planning a shoot, especially for apparel, footwear, grocery, jewelry, and regulated products.
A strong Amazon listing gives every image a different job
Amazon recommends at least six product images and one video, while the main image remains the only image shown first in search and browse surfaces. The recommendation appears in Amazon's updated Product Image Requirements summary. Amazon also describes three essential image roles: the product on white, the product in an environment, and product information such as dimensions or nutritional facts.
A useful six-image plan is more important than filling slots with near-duplicates:
| Image role | What it should answer | Typical treatment |
|---|---|---|
| Main | What exactly is being sold? | Actual product, pure white, no added copy |
| Alternate angle | What does another side look like? | Clean product view |
| Back or underside | Are there ports, closures, labels, or construction details? | Clean product view |
| Detail | What material or feature deserves inspection? | Tight crop with enough resolution |
| In context | How is the product used, worn, or scaled? | Lifestyle scene or model image |
| Product information | What dimensions, ingredients, or included parts matter? | Secondary infographic or annotated image |
Secondary images can carry context that would make the main image non-compliant, but they must still represent the product accurately. Amazon's own A+ Content design guidance recommends clear, well-lit product and lifestyle photos, multiple angles, and models for products such as clothing and accessories.
AI can help draft alternate views, but each view still needs to match the physical product. The AI camera-angle control guide explains the difference between controlled framing and asking a generator to invent a side it has never seen.
What changed for AI-generated Amazon product images in 2026?
Amazon now requires a metadata tag on product media that contains a photorealistic person generated entirely by AI. In its 2026 seller announcement, Amazon instructs sellers to use an IPTC-compatible metadata editor and add contains-synthetic-performer to the dc:subject XMP field before uploading the file.
The tag is not required when an image contains no people, contains only non-photorealistic people, or shows a real person who was altered with AI. Amazon says it will add a customer-facing indicator where applicable. Add the metadata after the last edit or export, then confirm that the final upload file still contains it.
This disclosure rule is not blanket approval for every generated image. Amazon still requires the image to accurately represent the product, and the main image still has stricter rules against mockups, placeholders, and misleading additions. The conservative approach is to anchor any AI-assisted work in real Product Photos, compare every output against the physical item, and keep a real product capture available as the main-image fallback. For a focused policy answer, read whether Amazon allows AI-generated product images.
How much does Amazon product photography cost in 2026?
Amazon product photography costs depend more on image type and review work than on the marketplace itself. DIY has low cash cost but consumes the seller's time; remote studios commonly price each delivered image; AI tools use subscriptions or Credits and still require product review.
| Production path | Current cost shape | Costs people often miss | Best fit |
|---|---|---|---|
| DIY | Equipment, software, and your time | Reshoots, clipping, color correction, learning | A small, stable set of simple products |
| Remote studio | Per image, plus selected services | Shipping, styling, models, premium retouching, revisions | Exact hero imagery and difficult materials |
| AI product photography | Subscription or Credit allowance | Source preparation, curation, product checking, metadata | Repeat secondary imagery and catalog-scale production |
| Hybrid | Studio or DIY main image plus AI-assisted secondary set | Managing two approval paths | Catalogs that need both exact capture and repeatable variation |
Current provider pricing gives a grounded benchmark rather than a universal market average. ProductPhotography.com lists seamless-white photos from $33 each, hero images from $115, and lifestyle images from $195. Soona's current pricing lists $39 per photo plus a $149 studio pass for a one-off booking, with models, styling, and some edits priced separately.
Consider a 50-product catalog with five seamless-white images and one lifestyle image per product. At ProductPhotography.com's published starting rates, the calculation is 50 × ((5 × $33) + $195) = $18,000 before shipping or add-ons. A simpler 300-photo Soona booking is 300 × $39 + $149 = $11,849 before paid upgrades. These are planning floors based on two providers, not quotes for every product or creative brief.
Nightjar uses a Subscription with Team-owned Credits rather than a per-shoot quote. A completed 1K or 2K Product Photography Single shot normally costs one Credit, a completed 4K output costs two Credits, and a four-image Photoshoot costs two Credits total. The more important economic difference is reusable setup: Products retain the subject evidence, while Recipes retain the production direction, so the work invested in one shoot can carry into the next product instead of ending with the exported files. The AI-versus-studio cost guide covers the broader fixed and variable cost model.
The best Amazon product photography workflow is usually hybrid
The best Amazon product photography approach is usually hybrid because the main and secondary images solve different problems. Amazon's current guidance combines strict main-image rules with a recommendation for at least six images and one video, so one production method rarely gives both the strongest product verification and the most efficient variation. Use the method with the strongest verification for the main image and the method with the best repeatability for secondary images; catalog size matters, but product risk matters more than an arbitrary SKU threshold.
| Seller situation | Best starting approach | Reason |
|---|---|---|
| Few simple products and available time | DIY main and secondary set | The setup can be learned once and reused without a large cash commitment |
| Reflective, transparent, regulated, color-critical, or mechanically detailed product | Professional studio | On-set control and specialist retouching are worth the cost when a small visual error changes the product claim |
| Recurring launches across many products | Product-photography AI plus human approval | Reusable direction reduces re-briefing and catalog drift |
| High-stakes hero product with many routine variants | Hybrid | Exact capture establishes the source of truth; AI extends the approved visual system |
| Photorealistic AI-generated person | AI plus metadata workflow | The final file needs Amazon's synthetic-performer tag before upload |
Professional photography remains the safer choice when color, texture, readable labels, scale, reflectivity, or functional details must be represented with tight art direction. AI is most useful when a seller has sound source images and needs repeatable lifestyle scenes, model imagery, crops, or cohesive secondary variants. DIY makes sense when time is available and the lighting problem is simple.
If AI is the right production path, compare tools on product evidence, repeatable direction, review behavior, and output controls rather than the cheapest generated file. The AI product photography tools guide uses those criteria across the current field.
Repeatable Amazon secondary images need two kinds of memory
Repeatable Amazon secondary imagery needs two kinds of memory: evidence of what the product is and reusable direction for how it should look. Nightjar implements that split with a Product that groups multiple Product Photos, a factual description, and optional physical dimensions, while a Recipe saves the Photography Style, background choice, Framing or Pose, and output settings. The practical result is that an approved Amazon gallery direction can move across a catalog without asking the generator or the next team member to reconstruct it.
For a clean product-only image, Product Photography can use a flat white background with a Framing choice and a Shadow setting. For a lifestyle secondary image, the same Product can use a saved Background and Photography Style. When a person appears, a reusable Fashion Model and Pose plus Camera Distance carry the model identity, body arrangement, and crop. A Photoshoot is the four-image output choice inside Product Photography; it creates varied but connected images from one direction rather than acting as a separate Workflow.
Built-in review is a second line of defense, not the seller's sign-off. Nightjar can use multiple Product Photos and built-in visual review to catch and retry obvious eligible failures at no extra Credit cost, but it does not guarantee Amazon acceptance or perfect product reproduction. Check the finished product shape, color, materials, text, logos, included parts, background, crop, resolution, and any required AI-person metadata before publishing. The help desk explains how to keep a catalog on one background, how to build a lifestyle scene from a white-background photo, and how to prevent the product's shape from changing in a new scene.
What should an Amazon image quality-control checklist include?
An Amazon image quality-control checklist should test the exported file, the product claim, and the applicable category rules before upload. A generated or retouched image can look convincing and still fail one of those tests.
- Open the current Amazon Product Image Guide and the product's category style guide.
- Confirm the longest side is within 500 to 10,000 pixels; aim for 1,600 pixels or more.
- Sample the main image background and confirm pure white at RGB 255, 255, 255.
- Check that the actual product fills at least 85% of the main frame without being cropped.
- Remove added copy, badges, graphics, borders, watermarks, and unincluded props from the main image.
- Compare color, shape, count, packaging, labels, text, dimensions, and included parts against the physical product.
- Add
contains-synthetic-performermetadata if the file contains a photorealistic person generated entirely by AI. - Upload through Seller Central, inspect the live detail page, and use Image Manager or Seller Support if an accepted contribution is not displayed.
Do not outsource the last review to a generator, studio, or editing app. Amazon decides whether an image is accepted and displayed, and category rules can change after a production workflow is set up.
Frequently Asked Questions
What size should Amazon product images be in 2026? Amazon accepts 500 to 10,000 pixels on the longest side and prefers files above 1,000 pixels for zoom. Amazon's 2026 guidance calls 1,600 pixels or more optimal, so 1,600 is the practical target.
Does an Amazon main image need a pure white background? Yes. The general main-image rule specifies pure white with RGB values 255, 255, 255. Check the final export, because shadows, masks, and compression can leave off-white edge pixels.
Can Amazon sellers use AI-generated product photos?
Amazon's current rules contemplate AI-generated media, but every image must still represent the product accurately and meet the relevant main, secondary, and category requirements. Images containing photorealistic people generated entirely by AI also need Amazon's contains-synthetic-performer XMP metadata tag.
How many images should an Amazon listing include? Amazon currently recommends at least six product images and one video. Use distinct roles, such as a compliant main image, alternate views, a detail, a lifestyle image, and product information, instead of repeating the same view.
Can an Amazon secondary image contain text or an infographic? Secondary imagery can communicate dimensions, ingredients, usage, or other product information that does not belong on the main image. The claim and graphic must remain accurate, legible, and consistent with the listing and category rules.
Do Amazon sellers need a professional product photographer? Not for every image. Professional photography is the strongest choice for high-stakes or technically difficult products, while DIY or AI-assisted production can handle routine catalog and secondary-image work when a person reviews every export.
Why was an Amazon product image rejected or suppressed? Common causes include an off-white main background, insufficient product coverage, added text or watermarks, props that imply something is included, low resolution, visible artifacts, inaccurate product representation, or a category-specific violation. Seller Central's Image Manager and the rejection notice are the first places to check.
References
- Amazon Product Image Guide - Official baseline and category-specific image rules
- Amazon Product Image Requirements summary - File limits, zoom guidance, recommended image count, and enforcement
- Amazon 2026 product photography guidance - Current 1,600-pixel production recommendation and main-image checklist
- Amazon AI-generated person metadata announcement -
contains-synthetic-performertagging requirement - Amazon A+ Content design guide - Product, lifestyle, angle, and model imagery guidance
- ProductPhotography.com pricing - Current seamless-white, hero, and lifestyle starting prices
- ProductPhotography.com process - Delivery, shipping, revisions, and per-photo pricing model
- Soona pricing - Current per-photo, studio-pass, and add-on pricing
- Nightjar - AI product photography system