If you sell on Amazon, you already know the game: your product listing lives or dies on the images. Traffic without conversions is expensive. And conversions come from photos that pass Amazon's rules and actually make shoppers want to buy.
For years, the only way to get Amazon-ready photos was to ship your product to a studio, wait a week, pay ₹8,000–₹25,000, and hope the shots came back usable. In 2026, that's no longer the fastest — or even the cheapest — path.
This is the full playbook for creating every image slot on your Amazon listing with AI. Real models, real prompts, real costs.
Why Amazon photography is the hardest photography problem
Amazon isn't Instagram. You can't cheat with a filter and call it done. Every image has to:
- Pass Amazon's technical rules. The main image has to be on pure white (RGB 255, 255, 255), the product has to fill 85% of the frame, no props, no text, no watermarks, no additional objects. Anything else and Amazon can suppress your listing.
- Compete visually. You're up against 30–50 other sellers of nearly identical products. Same category, same price band, same keywords. The photos are the differentiator.
- Cover 7 different slots — hero, infographic, lifestyle-in-use, dimensions, detail close-up, environmental scene, brand story. Each slot has a different job. Each slot needs a different treatment.
- Look consistent as a set. Buyers scroll through all 7 in about 4 seconds. If they feel like they were shot in 7 different studios by 7 different teams, trust drops and the conversion rate drops with it.
Traditional studio shoots handle 1–3 of those slots well and outsource the rest to graphic designers who bolt on infographic text after the fact. It's slow, expensive, and inconsistent.
AI, done right, handles all 7 in one workflow.
The 7 Amazon listing image slots (and what each one is for)
Amazon lets you upload up to 9 images. The first 7 are the ones that matter. Here's what each slot needs to accomplish:
- Hero (main image) — Pure white background. Product fills 85% of frame. This is the image Amazon shows in search results, so it has to work at thumbnail size. Amazon's white-background rule is enforced on this slot specifically.
- Feature infographic — Product on clean background with 3–5 text callouts highlighting the features that matter most. This is where buyers scan to decide whether to keep scrolling.
- Lifestyle-in-use — A person actively using the product in a natural, real-world setting. Buyers need to picture themselves owning it.
- Dimension spec — Clean product shot showing size and form. Often overlaid with measurement callouts in the final version.
- Detail close-up — Macro shot of texture, material, craftsmanship. Communicates quality and premium feel.
- Environmental scene — Product in a beautiful contextual environment, no people. Aspirational and lifestyle-driven.
- Brand story or USP highlight — A bold headline claim (e.g. "Waterproof to 30m") overlaid on a clean product image. This is where you close the sale.
Each of these slots has different requirements — and this is where AI product photography workflows have gotten dramatically better in the last year.
The AI models that actually work for Amazon photos in 2026
Not every AI image model handles Amazon-quality product work. The three that do:
Nano Banana 2 (Edit endpoint) — Google's Gemini-based image editor. Best for lifestyle scenes with people holding or using your product. It reads your product as a locked visual reference and builds a natural, grounded scene around it. This is the model that produces the "person on a Mumbai street holding your bag" shots that convert.
Flux Kontext Pro / Max — Purpose-built for product preservation. When you need to change the background completely while keeping the product's exact shape, colors, labels, and branding, this is the model. Kontext Max has slightly stronger prompt adherence for complex scene instructions.
Flux Dev image-to-image — Adjustable strength (typically 0.85) gives you creative freedom while still recognizably preserving your product. Good for detail close-ups and moody hero shots.
We deliberately avoid running Amazon hero shots through Bria's product-shot endpoint, even though it's marketed for products. In practice, Bria ignores the "white background" instruction about 40% of the time and generates its own scenes. Nano Banana 2 Edit with a strict prompt about RGB(255,255,255) is far more reliable for hero compliance.
The 7-slot workflow, mapped to models
Here's the exact model-per-slot mapping we recommend after testing thousands of products:
| Slot | Model | Why |
|---|---|---|
| 1. Hero (white BG) | Nano Banana 2 Edit | Reliable white background compliance, product fills frame |
| 2. Feature infographic | Flux Kontext Pro | Clean background, then overlay text via Sharp |
| 3. Lifestyle in use | Nano Banana 2 Edit | People + product scenes look natural, not composited |
| 4. Dimension spec | Flux Kontext Pro | Clean white base, dimensions overlaid programmatically |
| 5. Detail close-up | Flux Kontext Max | Best texture and material detail preservation |
| 6. Environmental scene | Nano Banana 2 Edit | Aspirational scenes without people |
| 7. Brand story / USP | Flux Kontext Pro | Clean product for headline overlay |
For text overlays on slots 2, 4, and 7 — the feature callouts, dimension labels, and USP headlines — don't let AI generate the text. AI-generated text is still unreliable at listing scale. Generate the clean image with AI, then composite crisp SVG text on top programmatically. This is how you get infographic-quality text without the classic "AI wrote 'Featues'" typos.
Writing prompts that actually work for Amazon
Prompt structure matters more than length. The pattern that consistently works:
[Purpose] + [Product context] + [Scene/lighting] + [Compliance rules] + [Quality tags]
For the hero slot, a prompt that works reliably:
Professional Amazon product photography. Pure white background exactly RGB(255,255,255). Product centered and filling 85% of the frame. Soft even studio lighting from directly above, no shadows on background, no colored surfaces, no props, no text, no watermarks. Clean commercial e-commerce photography. Square 1:1 format. Ultra sharp product details.
Notice the specifics: exact RGB values, exact fill percentage, explicit "no props / no text / no watermarks" — the compliance rules that Amazon actually enforces. Vague prompts like "white background product photo" give you inconsistent results.
For the lifestyle slot, structure changes:
Lifestyle product photography. A person actively using [product] in a natural real-world [category] setting. [USP headline]. Natural lighting, authentic candid moment, person clearly interacting with the product, product fully visible and prominent, photorealistic commercial photography.
The word "actively using" and "clearly interacting" are the specific phrases that produce the shot where the product is prominent. Without them you get shots where the person overshadows the product.
The cost comparison, honestly
Let's do the actual math for a single Amazon listing.
Traditional photography approach:
- Studio hire + photographer: ₹12,000–₹25,000
- Product courier round-trip: ₹500–₹1,500
- Graphic designer for infographic overlays: ₹3,000–₹6,000
- Turnaround: 5–10 business days
- Total per listing: ₹15,500–₹32,500
AI approach (Oranokai or similar):
- 7 images × ₹6–₹17 per image: ₹42–₹119
- Text overlay compositing: included
- Turnaround: 30–90 seconds per image, ~10 minutes for the full 7-slot set
- Total per listing: ~₹120
That's a 99% cost reduction and a 500× speed improvement. Even factoring in the tool subscription, the break-even hits after a single listing.
For sellers with 20+ products, the numbers become unsustainable to argue with. A brand launching 100 SKUs a year saves ₹15–30 lakh annually just on photography.
Common Amazon photography mistakes (that AI won't fix on its own)
AI does not fix bad inputs. Some things you still need to get right yourself:
1. The source product photo has to be clean. Blurry phone photos taken in bad lighting produce blurry outputs. Take your source photo with decent light, in focus, and with a reasonably plain background. AI enhances — it doesn't hallucinate detail that isn't there.
2. Don't over-prompt. Long prompts with 15+ style descriptors dilute the important instructions. Keep prompts under 60 words. Put compliance rules (white background, no text, product fills frame) at the front.
3. Match the aspect ratio to the slot. Amazon's hero image is displayed square. Generating a portrait image and cropping it produces awkward crops. Generate at 1:1 for hero and dimension slots, then use appropriate ratios for others.
4. Don't skip the detail close-up. Sellers routinely skip slot 5 because it feels less important. Data says the opposite — buyers who click through to detail images convert 2.3× higher. Generate a good macro shot.
5. Keep the set visually consistent. All 7 images should feel like they came from one campaign. Use the same product image as the reference across all slots so lighting, color, and framing stay coherent.
Getting started in 4 steps
If you want to move fast:
- Take one clean source photo of your product. Phone camera, natural window light, plain background. That's your reference.
- Sign up for Oranokai (or another AI product photography tool with an Amazon workflow).
- Run the 7-slot workflow. Upload your source image, add the product name and 3 key features, hit generate. All 7 slots produce in parallel in about 8–10 minutes.
- Upload directly to Amazon Seller Central. Full-resolution downloads at 2K are Amazon-listing-ready without further editing.
You'll spend more time uploading to Amazon than you did generating the photos.
The bigger picture
Amazon product photography used to be a bottleneck. Every new SKU meant a photoshoot, a designer, a delay. That bottleneck defined how fast a brand could grow.
In 2026, the bottleneck is your imagination. If you can describe the scene and provide a source photo, you can generate professional-grade Amazon listing images in the time it takes to make coffee.
The sellers who understand this — and act on it — are the ones stacking listings, iterating on their creative, and outrunning slower competitors on both cost and speed.
Ready to build your Amazon listing set? Try Oranokai's Amazon workflow free →
Oranokai is an AI-powered product photography platform built for Amazon sellers, D2C brands, and 3D artists. Generate all 7 Amazon listing slots — hero, infographic, lifestyle, dimensions, detail, environmental, and USP — with a single upload. Powered by Nano Banana 2, Flux Kontext, and Flux Dev image models.