Generator
AI Product Photo Generator on Astorie
Upload a photograph of your product, choose a backdrop or a scene, and generate new product images: white-seamless packshots, lifestyle shots, flat lays, and packaging mockups. The uploaded photograph is used as a reference, which helps keep labels, materials, and packaging readable while the setting changes. On the Astorie canvas the approved image can then be carried into a video ad without being uploaded again.
Free to start. No card required.
Reshoot a product photo right now
Drop in a product shot, pick the set you want it photographed on, and hit Create. The same generation pipeline the canvas uses relights the product on a new background, on a real project you can keep editing.
Before and after examples
Before
AfterBefore
AfterWhat you can generate
- E-commerce hero shots on white seamless or brand-color backdrops
- Lifestyle scenes in kitchen, bathroom, vanity, or desk environments
- Packaging mockups before the print run is committed
- Brand-color hero compositions with matching backdrop and rim light
- Top-down flat-lay arrangements for skincare, food, and accessories
- Macro detail and texture close-ups for material storytelling
- Amazon listing packshots aligned to marketplace size specs
- PDP hero frames already framed for pan-and-zoom motion
- App store screenshots and product UI shots aligned to the brand style
Prompt examples
Beverage can on white seamless, soft front key plus rim light from camera right, brand-orange backdrop fade, deep contact shadow, 4:5 PDP framing.
White-seamless hero for marketplace listings. Keeps the can label crisp and centers the product for pan-and-zoom motion.
Beauty serum dropper bottle on marble vanity, morning daylight from camera left, soft linen draped behind, condensation droplets, 4:5 lifestyle framing.
Lifestyle vanity scene that doubles as the establishing frame for a Seedance 2 macro reveal.
Sneaker on minimal concrete plinth, deep cool shadow, brand-blue accent gel from above, low camera angle, 16:9 hero framing.
Editorial sneaker packshot. The plinth gives the orbit motion in the next step a clean foreground.
Top-down flat-lay of a skincare set on warm linen, soft north-facing window light, gold leaf accents, even spacing, 1:1 social framing.
Flat-lay composition for an IG carousel. The symmetry survives resequencing without looking generated.
Smartphone retail packaging unboxing scene on walnut desk, sunset key from camera right, ambient fill, focus on packaging seam, 9:16 framing.
Vertical unboxing setup, framed in advance for an image-to-video Reels cut.
Candle on wooden side table next to open hardback book, warm tungsten side light, soft bokeh blanket in background, 4:5 framing, deep negative space top-right for headline overlay.
Cozy lifestyle frame that keeps headline space free for ad-creative iteration.
Macro material detail of leather wallet stitching, hard rim light, neutral grey backdrop, shallow depth of field, 1:1 framing for detail card.
Texture-led close-up for the detail-loop step in a multi-shot product cut.
Packaging mockup of a 500ml glass bottle with custom label, studio softbox lighting from front-left, neutral seamless, 1:1 framing for label proof.
Pre-print mockup. Surface label drift before the print run, then approve into the video pipeline.
What you need to know
- Accepted files
- JPEG, PNG and WebP. Photos in HEIC or HEIF are converted to JPEG in your browser before they upload. GIF, animated WebP and SVG are not accepted.
- Image size
- Both sides must be at least 300 px. PNG and WebP files must be 30 MB or smaller. Larger JPEG files, and very large images of any accepted type, are resized in your browser before they upload.
- Uploading before you sign in
- You can attach an image before you have an account. It stays in this browser, up to 25 MB, and is uploaded only when you start a generation.
- What you get back
- One PNG image at 1K resolution, in a 1:1 frame.
- Batch generation
- This tool runs one generation at a time, and each run returns a single result.
- What it costs
- Uses up to 20 credits per image. Plan discounts and included allowances may lower this.
- Free allowance
- A new account starts on the free plan with 40 credits a month. Some accounts are asked to claim that allowance once, on the credits page, before it can be spent. See pricing for what the paid plans include.
- Your files and results
- Uploads and generated results belong to the project they run in. This demo creates that project in your own account rather than in a team workspace. How long they are kept, and how to have them deleted, is set out in the privacy policy.
- Commercial use
- What you may do with a result is governed by the terms, and the model you pick may carry its own restrictions. Read both before you publish or sell a result.
Best Astorie workflow
Why this is more than a one-shot generator on Astorie.
- The product photograph, the brand-color reference, and the scene reference stay as separate inputs on the canvas rather than being folded into one prompt, which helps the model hold the product detail while the setting changes.
- Hero placements can run across Imagen 4, Nano Banana 2, Flux Kontext, and Seedream in parallel so the strongest result wins per placement, and the rest of the catalog runs on whichever setup won.
- The approved still is the input to the next step. A Seedance 2 hero spin or a Runway Gen-4 lifestyle cutaway starts from the same image node, and the 1:1, 9:16, and 16:9 cutdowns export to Premiere Pro or DaVinci Resolve.
- Run a cleanup pass on GPT Image 2 or an upscale before the video step. Artifacts that are barely visible in the still become obvious once the image moves, and catalog delivery usually asks for 4K masters.
- Save the finished setup once it works. The next launch changes the product photograph and the brand-color reference; the lighting, framing, and motion stay where they are.
Where to take the output next
Follow /how-to/create-product-photos for the per-model walkthroughs behind the recommended stack.
Send the approved frame into /features/ai-product-video-generator to become the campaign cut without re-uploading the product image.
Hand the bundle to /features/ai-video-nle-export for clean 1:1, 9:16, and 16:9 versioning into Premiere or DaVinci.
Pair with /prompts/image/product-photography-prompts for paste-ready studio and lifestyle recipes.
Recommended models
nano-banana-2
imageStrongest reference adherence to product detail: labels, materials, and packaging text usually survive a scene swap.
imagen-4
imagePhotoreal premium hero placements with natural-light fidelity for beverage, beauty, and tech SKUs.
flux-kontext
imageEdit-aware product placement into new scenes while preserving the original packaging detail.
seedream
imageEditorial and lifestyle range when the brief calls for mood-led PDP frames.
gpt-image-2
imageCleanup pass for fine label legibility, color contrast, and edge polish before the video chain.
Frequently asked questions
Do I need to upload a photo of my product?
For anything that has to match a real SKU, yes. The uploaded photograph is what the model reads for shape, material, and label, and it is the difference between a picture of a product and a picture of your product. Text-only prompts are fine for concept and mood work where nothing has to match an existing item.
How do I stop label and packaging text from drifting?
Use Nano Banana 2 as the upstream model, feed the brand-approved photograph as a reference rather than describing the label in the prompt, and run a cleanup pass on GPT Image 2 before export. Re-describing the label fights the reference and is the most common cause of drift. Review each output against the source before locking it, particularly label copy and material highlights.
Which image formats can I upload?
JPG, PNG, and WebP upload directly, and HEIC or HEIF photos are converted to JPEG in the browser first. GIF, animated WebP, and SVG are not accepted. Both sides have to be at least 300 pixels. The 30MB ceiling is a hard limit for PNG and WebP only, which are rejected above it rather than recompressed; larger JPEGs are re-encoded in the browser, and very large images of any accepted type are downscaled before upload.
What format is the download, and is it big enough for print?
The image downloads in the format the model produced, usually PNG or JPEG, and Astorie streams the original file rather than re-encoding it. Generation resolution depends on the model, so chain the selected masters through image upscale when the delivery spec asks for 4K or print resolution.
Can I generate Amazon-spec packshots?
Generate at the marketplace size, then check the cleanup pass for white-seamless RGB compliance and product-frame ratio before submitting. Marketplace specifications change, so confirm the current requirements against the marketplace itself rather than against this page.
Can I use the images commercially?
Astorie does not claim ownership of the images you generate. Whether a specific result can be used commercially also depends on the rights in the product photograph and any reference images you uploaded and on the terms of the model that produced it. Check the Astorie terms of service before publishing to a store, a marketplace, or paid placement.
Are my product photos private?
Uploads and generated results are stored against your account. How they are processed, how long they are kept, and how to delete them are set out in the Astorie privacy policy, which is the authoritative source; this page does not restate its terms. It is worth reading before uploading unreleased packaging.
Is there a free allowance?
Astorie has a free plan that does not require a credit card and comes with a monthly credit allowance. Some accounts are asked to claim that allowance once, on the credits page, before it can be spent. Cost per image varies by model and output size, so the current allowance and per-model pricing are listed on the Astorie pricing page.
Can I generate a whole catalog in one pass?
Most image models can return several variations per generation, and a group of nodes on the canvas can be run as a single batch, which is how a set of SKUs goes through the same setup. The number of variations per run depends on the model. Test one SKU before running the batch; a setup that is slightly off is cheaper to fix on one product than on forty.
Can AI replace a real product photo shoot?
For catalog volume, lifestyle variants, and quick iteration it covers a lot of the work at a fraction of the time and cost. For top-tier campaign hero shots a real shoot still tends to win on shadow, material, and skin detail. Treat it as augmentation rather than replacement, with the split depending on how much the brand rests on photography.
Do I need a separate tool for the video version?
No. The product photo node wires into Seedance 2, Runway Gen-4, and Kling 3 video nodes on the same canvas. See /workflows/ai-product-video for the end-to-end pipeline.
Related features
Related how-to guides
Related prompts
Related workflows
Related reading
Generate it on the canvas
Open Astorie, drop this generator on the canvas, and wire it into the workflow you actually need. Free to start — no card required.