Google Shopping image requirements: size, rules and AI tags
Google Shopping images must be at least 500 × 500 px from January 31, 2027, under 64 MP and 16 MB, with no promo overlays; AI images need AI metadata.

Google Shopping product images must be at least 500 × 500 pixels (for all products from January 31, 2027), no larger than 64 megapixels or 16 MB, and show the whole product with no promotional overlays such as watermarks, prices or logos; Google recommends around 1500 × 1500 pixels or above. An image made with generative AI must also carry metadata saying it was AI-generated, such as the IPTC DigitalSourceType TrainedAlgorithmicMedia tag.
Google's rules are quoted from the Merchant Center Help page [Image link [image_link]](https://support.google.com/merchants/answer/6324350?hl=en); Sume facts come from the Image API docs and the Sume API reference. All were read on 2026-09-29.
What are the Google Merchant Center image requirements?
Every product needs a main image in the image_link attribute; other views go in additional_image_link. Google disapproves a product whose image misses a minimum requirement and tells you in the Diagnostics section of Merchant Center. Google says the 500 × 500 size requirement applies to all products beginning January 31, 2027.
| Requirement | What Google says |
|---|---|
| Size | At least 500 × 500 px; around 1500 × 1500 px or above recommended |
| Largest image | No image larger than 64 megapixels or 16 MB |
| File formats | JPEG, WebP, PNG, GIF, BMP and TIFF |
| Product | Show the entire product, with minimal or no staging |
| No promotional elements | No calls to action, prices, free shipping, watermarks, logo overlays, barcodes or borders |
| AI-generated images | Must contain metadata saying the image was AI-generated, such as IPTC DigitalSourceType TrainedAlgorithmicMedia |
Can I use AI-generated product images on Google Shopping?
Google's page sets a rule for them rather than a ban: all images created using generative AI must contain metadata indicating that the image was AI-generated, and you must not remove embedded tags such as the IPTC DigitalSourceType property. The main image must still accurately show the entire product, and Google doesn't allow a generic image or illustration in place of the actual product, outside a few named categories.
Sume's docs don't say whether generated files carry IPTC or other provenance metadata. Before you submit, open each file in a metadata tool, check for the tag, and add it if it's missing.
What does Google recommend beyond the minimum?
- Use a solid white or transparent background; staged or lifestyle images that clearly show the product are also acceptable. Don't use a transparent background for light-colored products, which might show on black.
- Frame the product to take up no less than 75% and no more than 90% of the image.
- Don't scale up an image or submit a thumbnail; submit the largest, highest-resolution image you have.
- Use a stable URL without timestamps or parts that change, and a new URL when you replace an image.
How do I make a Google Shopping image at the right size?
Render it at the size you need rather than upscaling, since Google advises against scaled-up images. ChatGPT Image 2.5 (openai/gpt-image-2.5) on Sume takes an exact custom image_size with both edges multiples of 16, so 1504 × 1504 is the nearest square at or above Google's recommendation. Put your real product photo at a public HTTPS URL in input_references so the image starts from the actual product, and check that the result still shows it accurately.
The result comes back as a signed, Sume-hosted URL in data[].url. Don't put that URL in your feed: download the file and host it at a stable URL of your own, as Google asks. For a white-background cutout of a real product photo, see White background product photo with AI.
Sources
- [Google Merchant Center Help: Image link [image_link] (read 2026-09-29)](https://support.google.com/merchants/answer/6324350?hl=en)
- Image API
- Sume API reference
- API reference
Related posts
More in Media tools
- AI video HDR and EXR export: what Sume's outputs are
Luma says Ray3.2 has native HDR with 16-bit EXR export. Sume's docs describe MP4 outputs and no HDR or EXR; here is where that matters in a finishing pipeline.
- How to flip a video horizontally with an API
To mirror a video left to right, run the hflip filter. Sume's video filter takes it as a one-line filtergraph and returns a new MP4.
- How to make a video montage: clips cut to one music track
To make a video montage, pick the music first, mark where the cuts go, then lay short clips over the track at those times in one render.
- Can AI make a song from a voice memo or a video?
Suno v6 says it can start from a voice memo, image or video. Sume's music API takes a text prompt and one optional image, so audio and video stay outside it.
Written by Sume