---
title: "AI product photos for Amazon listings | Herofab"
description: "Upload a phone photo of a product. Herofab delivers it alone on white, and the download sized for Amazon brings the background to pure white."
url: https://herofab.app/tools/product-photos-for-amazon
language: en
updated: 2026-11-20
---

# Amazon main images, pure white and whole in frame.

Amazon's main image shows the whole product on a pure white background, filling most of the picture with nothing cut off. Herofab's packshot studio makes that image from a phone photo taken on any surface, and when the result is sized for Amazon its background is brought to pure white.

[Start free](https://app.herofab.app/login?lang=en&entry_source=tool) · [Packshot studio](https://herofab.app/tools/packshot-studio)

![The brown leather tote alone on a white studio background, front view · Packshot studio](https://herofab.app/img/ex/tote-packshot-1100.webp)

Made with Herofab · Packshot studio

## What Amazon measures in a main image.

1. ### Pure white, not near white

   Amazon names a single background for the main image: pure white. A studio white that looks clean on a screen can still sit a shade below it. The packshot studio delivers an evenly lit white, and the download sized for Amazon brings that background to pure white, so the Amazon file is the one to use as the main image.
2. ### Large in the frame, nothing cut off

   The product has to fill most of the image and stay entirely inside it. A strap, a heel or a handle that runs out of the picture is a cut-off edge. Every packshot shows the product once, centred and filling most of the frame; a long strap spread wide makes the product itself small, so look at the margins before uploading.
3. ### Sharp pixels, and more than one image

   Amazon publishes a minimum and a maximum for the longest side, and the image has to be clear, with no pixelation or jagged edges. The Amazon download sits inside that range. One image per product is the minimum and Amazon recommends several more, so further angles from the same source photo are worth planning.

## What Amazon asks for.

Source: [Amazon — Sell on Amazon: product photos](https://sell.amazon.com/blog/product-photos). Checked 10 October 2026.

- **Size:** 500 to 10,000 pixels on the longest side
- **Formats:** JPEG, TIFF, PNG or non-animated GIF
- **Quality:** Clear, unpixellated, with no jagged edges
- **Number:** At least one image per product; at least six are recommended
- **Product fill:** Recommended: the product fills 85% or more of the frame (Amazon's shooting advice)
- **Background:** The main image must have a pure white background (RGB 255, 255, 255)

[Amazon product image requirements](https://herofab.app/guides/amazon-product-image-size) · [Check your images — free](https://herofab.app/tools/amazon-product-photo-checker)

## The source photo behind an Amazon main image.

- Keep the entire product inside the photo, straps and handles included; a part outside the frame cannot be added back faithfully, and Amazon wants nothing cut off.
- Arrange the product compactly, a strap tucked in or laid close to the body, so that it can fill most of a square image instead of leaving wide empty margins.
- Focus on the product and use even daylight: Amazon's zoom enlarges the picture, and a soft label or a blurred line of stitching is what a shopper sees first when zooming in.

## Soft outlines, pixelation and the wrong file type

Beyond the white ground and the fill rule in the table, image quality is a reason in itself: a picture should be clear, without pixelation and without jagged edges. Jagged outlines often come from a hard cut-out made at low resolution, and pixelation from a small supplier photo stretched to reach the zoom threshold.

File type is the other quiet reason. JPEG is the preferred type, PNG and TIFF are accepted, and GIF only without animation. The white-ground rule is written for product shots in most cases, which leaves room for scenes in later slots but not in the main one, where the whole product has to stay inside the frame.

## Packshots for objects, an upright size for garments

Objects go through the packshot studio: the product alone and centred, rendered again under studio light, so its outline is drawn rather than cut and carries no fringe of the old background. Garments take Ghost mannequin, and the download menu offers a clothing entry in an upright three-by-four size beside the square for products.

Further slots come from the same upload: other packshot angles, a scene from the product studio, close-ups from Detail shot and worn views from Virtual try-on. An image showing a person Herofab generated carries the contains-synthetic-performer keyword in addition to the IPTC marker for AI media.

## What the download does when a product meets the border

Unlike most channel sizes, this download adds a safety margin: when a product on white reaches the edge of the delivered image, the file is built with a thin white border so that nothing is cut, and the backdrop is set to pure white in the same step. The menu then measures fill and edges and reports a shortfall rather than cropping to hide it.

A low fill is fixed at the source, with straps and handles arranged compactly or a new packshot, because enlarging the product would mean cutting it. The free checker for this marketplace measures background value, fill and edges in the browser, and the free resizer pads photos from other sources into the same square.

## Product photos for Amazon, answered.

**Is the file for Amazon different from the delivered image?**

Yes. After delivery, Amazon is chosen as the marketplace and a new file is made from the delivered image without cropping the product. Its size is within the limits Amazon publishes, and for a packshot the background is brought to pure white.

**What goes into the other image slots of an Amazon listing?**

Amazon recommends several images per product, and the white-background requirements on this page are written for the main one. The rest can come from the same upload: another packshot angle, a scene, or on-model images made with Virtual try-on. Images that show an AI-generated person carry the keyword Amazon asks for, and the models are synthetic, not photographed people.

**How is a photo tested against Amazon's rules before upload?**

Herofab's free Amazon checker reads the file in the browser and tests it against the published rules; the image is not uploaded. It does not replace the seller's own look at the product, small label text first, and it does not speak for Amazon, which decides whether a listing image is accepted.

**Why does a white background fail when it looks white on screen?**

The rule asks for an exact value, with red, green and blue all at their maximum, and a studio white can sit a few steps below it. The checker reads a band along the four edges and reports the colour it finds; the download for this marketplace sets an even off-white backdrop to pure white.

**Is there a separate image size for clothing listings?**

Yes. The download menu lists two entries for this marketplace: a square for products and an upright three-by-four size for clothing, taken from the clothing image guidance. Both bring an even backdrop to pure white and keep the whole garment inside the frame.

**How much does it cost?**

4 credits per image, plus one credit for 4K. In US dollars that is $0.16 to $0.48 per image before tax, from the largest plan to the smallest credit pack. The cost is shown before you generate, and if a job fails the credits are returned. The 20 free credits at sign-up cover 5 images with this tool.

[Packshot studio](https://herofab.app/tools/packshot-studio) · [Product studio](https://herofab.app/tools/product-photography) · [Main image](https://herofab.app/glossary/main-image)

Examples by product

[Product photos for Google Shopping](https://herofab.app/tools/product-photos-for-google-shopping)
