Is the AI product photo generator free?

Yes — 1 free credit on signup with Google, no card required. One credit covers the full pipeline: background removal plus scene generation plus storage of the result. After that, plans start at $19/month for 100 credits.

How does it compare to Pebblely or Mokker?

Pebblely is template-driven — you pick a curated background and the tool composites your cutout into it. Palmou AI uses Google Gemini 2.5 Flash Image (image-to-image) so each scene is generated end-to-end, with lighting and shadow that adapt to your actual product. Mokker is similar in approach but slower per image. Palmou's edge: under-15s generation and the 2026 SOTA model.

Will my product look the same in the generated scene?

Yes. Image-to-image generation preserves the product's exact shape, color, label text, logos, and texture. Only the background and the lighting on the surrounding scene change. This is the entire point of using a model like Gemini 2.5 Flash Image versus a generic text-to-image generator.

What product categories does it work for?

All e-commerce categories. We've tested jewelry, watches, sneakers, candles, food and beverage, art prints, cosmetics, electronics, skincare, clothing flat-lays, and homewares. Reflective metals, glass, and translucent packaging all handled correctly thanks to BRIA's 256-level alpha channel feeding into the scene step.

What's the output resolution?

Scene-generation jobs output around 1024 px on the longest side, sized for catalog listings, social ads, and most e-commerce platforms. WebP at quality 92 by default. If you need higher resolution for print, the background-removal-only flow preserves your source image's native resolution (so a 4K input gives a 4K cutout).

Under 15 seconds end-to-end. Roughly 6 seconds for background-removal-only, 10–15 seconds with a generated scene. Google Gemini 2.5 Flash Image is one of the fastest image-edit models available in 2026 — about 5× faster than Seedream 5 Lite or comparable models.

Can I use the images for Amazon, Shopify, Etsy, or paid ads?

Yes, on every plan including the free tier. Full commercial license, no attribution, no per-image cap after generation. The White Studio scene meets Amazon's main image requirement (RGB 255 white background).

Does it support custom prompts?

Not currently. Free-form prompts aren't exposed to keep output quality predictable — every scene ships server-side tuned. The six built-in scenes cover the most common e-commerce categories. If you need custom prompts at scale, the Agency plan REST API will eventually support them.

Free AI Tool

AI Product Photo & Image Generator

Upload one phone photo. Pick a scene. Get a catalog-ready product image in under 15 seconds. Powered by Google Gemini 2.5 Flash Image (nano-banana). No watermark, full commercial license.

Try free — no signup See how it works →

How AI product photo generation actually works in 2026

The 2024-era approach was to chain three tools: a background remover (U-2-Net or similar), a generative inpainter (SDXL + ControlNet), and a manual upscaler. Output was inconsistent, lighting was off, and the product silhouette warped on transparent or reflective surfaces.

The 2026 approach is single-pass image-to-image. Palmou AI feeds your source product photo and a curated scene prompt directly into Google Gemini 2.5 Flash Image (the model also called "nano-banana"). The model treats the entire input image as conditioning — your product silhouette, label text, logos, and color stay exactly as photographed, while the surrounding background regenerates to match the chosen scene. No precomputed mask, no inpainting bounds, no upscale step.

Six built-in scenes

White Studio
Pure-white seamless backdrop. Amazon-compliant for main listings.
Premium Beige
Warm cream-beige neutral. Universal across cosmetics, jewelry, fashion, food.
Nature
Lush green outdoor with bokeh. Lifestyle and clean-beauty positioning.
Luxury Marble
White Carrara marble with editorial lighting. Premium product positioning.
Urban
Concrete and street-style. Streetwear, sneakers, prosumer electronics.
Gradient
Pastel gradient backdrop. Modern minimalist e-commerce aesthetic.

Each scene is server-side prompt-engineered for product-photo fidelity. Custom prompts are intentionally not exposed today to keep output predictable.

What it's good for

Single-image catalog photos for online stores. Specifically tested and tuned for:

Jewelry

Chains, prongs, gemstones, reflective metals

Watches

Bracelets, dials, crystals, luxury positioning

Sneakers

Resale listings, StockX, GOAT, eBay

Skincare

Glass bottles, frosted finishes, label preservation

Cosmetics

Lipsticks, palettes, metallic packaging

Clothing flat-lays

Knitwear, fringe, sheer fabrics, hair

Candles

Glass jars, layered wax, seasonal scenes

Electronics

Glass screens, brushed metal, matte plastics

When NOT to use it

Print-grade ultra-high-res — scene output is ~1024 px. For billboard-print resolution, run the background-removal-only flow (preserves source resolution) and composite manually.
On-model fashion shots — Palmou AI is product-only. For garments on a model, use Booth.ai or a photographer.
Generic text-to-image — we only operate on an uploaded product photo. For ideation use ChatGPT, Midjourney, or DALL·E.

Frequently asked questions

Is the AI product photo generator free?: Yes — 1 free credit on signup with Google, no card required. One credit covers the full pipeline: background removal plus scene generation plus storage of the result. After that, plans start at $19/month for 100 credits.
How does it compare to Pebblely or Mokker?: Pebblely is template-driven — you pick a curated background and the tool composites your cutout into it. Palmou AI uses Google Gemini 2.5 Flash Image (image-to-image) so each scene is generated end-to-end, with lighting and shadow that adapt to your actual product. Mokker is similar in approach but slower per image. Palmou's edge: under-15s generation and the 2026 SOTA model.
Will my product look the same in the generated scene?: Yes. Image-to-image generation preserves the product's exact shape, color, label text, logos, and texture. Only the background and the lighting on the surrounding scene change. This is the entire point of using a model like Gemini 2.5 Flash Image versus a generic text-to-image generator.
What product categories does it work for?: All e-commerce categories. We've tested jewelry, watches, sneakers, candles, food and beverage, art prints, cosmetics, electronics, skincare, clothing flat-lays, and homewares. Reflective metals, glass, and translucent packaging all handled correctly thanks to BRIA's 256-level alpha channel feeding into the scene step.
What's the output resolution?: Scene-generation jobs output around 1024 px on the longest side, sized for catalog listings, social ads, and most e-commerce platforms. WebP at quality 92 by default. If you need higher resolution for print, the background-removal-only flow preserves your source image's native resolution (so a 4K input gives a 4K cutout).
How fast is it?: Under 15 seconds end-to-end. Roughly 6 seconds for background-removal-only, 10–15 seconds with a generated scene. Google Gemini 2.5 Flash Image is one of the fastest image-edit models available in 2026 — about 5× faster than Seedream 5 Lite or comparable models.
Can I use the images for Amazon, Shopify, Etsy, or paid ads?: Yes, on every plan including the free tier. Full commercial license, no attribution, no per-image cap after generation. The White Studio scene meets Amazon's main image requirement (RGB 255 white background).
Does it support custom prompts?: Not currently. Free-form prompts aren't exposed to keep output quality predictable — every scene ships server-side tuned. The six built-in scenes cover the most common e-commerce categories. If you need custom prompts at scale, the Agency plan REST API will eventually support them.

Ready to generate your first AI product photo?

Free credit on signup. No card. No watermark.