E-commerce product imagery
Turn a clean packshot into seasonal, lifestyle, or campaign-ready variants without rebuilding the photo shoot from scratch.
Loading…
Edit complete
Lime halo campaign
AI image editor
The AI image editor
for product photos.
One source, three campaign directions. Describe the edit and keep the product your campaign is built around.
ModePrompt edit
Keep the product. Build a charcoal studio campaign with a luminous acid-lime halo and reflective plinth.
Prompt-led Multi-image
AI image editing for commercial work
Edit an image with a direction, not a blank canvas.
An AI image editor changes a photo you already have from a written instruction. Avocado AI is an online AI image editor and AI photo editor for product photos and ads: upload a source image, describe the change, add references when the model supports them, and get new versions next to the original.
Creative teams and agencies
Explore more directions for a client asset, keep the original reference visible, and move from draft to a sharper creative brief faster.
Social and paid creative
Adapt the same core visual for new environments, colors, compositions, and references before you produce the final campaign assets.
AI photo editor, answered
What is Avocado's AI image editor?
Avocado's AI image editor changes a photo you already have with a written instruction. Upload a product shot, describe the edit, pick a model such as GPT Image 2.5 or Nano Banana 2, and get new versions next to the original, with commercial rights on every paid plan.
How do you edit a product photo with a prompt?
Upload the product photo, then write the change in plain words, for example "place the bottle on wet marble with soft morning light". Choose an edit model, generate, and compare the results with the source. Keep the best version or edit it again until the shot is ready for an ad or your store.
Which AI models can edit images in Avocado?
The editor lists GPT Image 2.5 Flare and Sunburst, GPT Image 2, Nano Banana 2, Nano Banana Pro, Seedream 5.0 Pro, Luma Uni-1 Max, Krea v2, Grok Imagine, Flux Kontext Max and Flux Kontext Lora. GPT Image 2.5 Flare is the default. You pick the model for each edit.
Can you edit with several reference images at once?
Yes. Multi-image models take extra images as references for style, texture, a second product or a scene. GPT Image 2.5 and Nano Banana 2 accept up to 14 images per edit, Seedream 5.0 Lite and Krea v2 up to 10, and Luma Uni-1 Max one source plus 8 references.
What is draw to edit?
Draw to edit lets you mark the part of an image you want changed by drawing on it, then describe the change. It makes the area and intent of an edit explicit, which helps when a prompt alone is not precise enough. Draw to edit is included from the Starter plan up.
How much does AI image editing cost?
Edits run on credits. Nano Banana 2 costs 1 credit per image at 1K or 2K and 2 at 4K. GPT Image 2.5 at 1K costs 1 credit at low quality and 4 at high. Editing is included from Starter, EUR 39 a month for 300 credits. You see the exact cost before you generate.
Avocado AI vs a typical AI photo editor
| Feature | Avocado AI image editor | Typical AI photo editor |
|---|---|---|
| Edit with a written prompt | Yes | Often |
| Choose the model for each edit | Yes, GPT Image 2.5, Nano Banana 2, Seedream 5.0 Pro and more | Usually one built-in model |
| Reference images per edit | Up to 14, depending on the model | Varies, often one image |
| Draw to edit | Yes, from Starter | Varies by tool |
| Turn the edit into a video ad | Yes, in the same workspace | Usually a separate tool |
| Repeat the edit for every product | Yes, with Flows | Usually manual |
| Commercial rights, no watermark | Every paid plan | Varies by plan |
How AI photo editing works
A practical editing workflow for product and campaign assets.
Start with the image you already have. The editor keeps the source asset, your instruction, and the resulting versions in one place so you can compare directions rather than starting over.
- 1
Upload the source image
Start with a product shot, visual reference, or creative asset you want to change.
- 2
Describe or draw the edit
Write the change you need. For more directed edits, use annotations and draw-to-edit tools where available.
- 3
Choose a model and iterate
Select an editing model based on the job, then generate and compare new versions against the original.
More than a background remover
Choose the right editing approach for the asset in front of you.
A fast cleanup, a product-scene variation, and a multi-reference creative direction do not need the same model. Avocado AI exposes the available edit models in one editor so you can choose the capability that fits the work.
Prompt edits
Describe scene, object, color, and styling changes in plain language.
Draw-to-edit
Use drawing and annotations to make the area and intent of an edit more explicit.
Reference-led edits
Add additional images when the selected model supports references for style, texture, or composition direction.
Continue your workflow
Generate, edit, and organize creative work in one place.
AI image editor FAQ
Questions about editing images with AI.
Is an AI image editor the same as an AI image generator?
No. An AI image generator makes a new picture from a text prompt. An AI image editor starts from a photo you already have and changes it, so your real product stays the anchor. Avocado has both: the image generator for new scenes and the editor for changing product shots you already own.
How do I use an online AI photo editor to create ad creatives?
Upload a clean product photo, describe the ad scene in the prompt, for example "protein tub on a gym bench in morning light", and generate a few versions. Keep the one that fits the ad, then turn it into a video ad in the video workbench or repeat the same edit for every product with Flows.
Can AI add an object or a second product to a photo?
Yes. Describe what to add and where, such as a towel next to the bottle or a second product on the shelf. For the closest match, add a photo of that item as a reference: GPT Image 2.5 and Nano Banana 2 take up to 14 images per edit. Compare the result with the original before you use it.
Where can I buy an AI photo editor for commercial use?
Avocado AI includes its AI photo editor in every plan from Starter, EUR 39 a month for 300 credits, billed monthly or yearly at 20% off. Every paid plan includes full commercial rights and no watermark, so edited images can run in paid ads, on product pages and in email.
Do I need a paid plan to use the AI photo editor?
Yes. Avocado runs on paid plans only. Plans start at EUR 19.99 a month for 100 credits, and editing, upscaling and draw to edit are included from Starter at EUR 39 a month for 300 credits. Yearly billing is 20% off, and credits roll over for up to a year.
Can I replace the background of a product photo?
Yes. Describe the new background in the prompt, such as a bathroom shelf, a gym bench or a plain studio sweep, and the edit model rebuilds the scene around your product. Generate a few versions, compare them with the original, and keep the one where the product looks right.
Will the edit keep my product's label and shape?
Edit models are built to keep the source image as the anchor, and results vary by model and prompt. Give a clear instruction, add a close-up of the label as a reference when the model supports it, and compare every version with the original before you use it in an ad.
Can I use edited images commercially?
Yes. Every paid plan includes full commercial rights and no watermark, so you can run edited images in paid ads, on product pages, in email and on social. Edited images are saved to your Avocado library with the rest of your work, ready to reuse in video or a Flow.
Can I turn an edited image into a video?
Yes. Open the edited image in the video workbench and use it as the start frame for image to video with models such as Seedance 2.5, Kling 3.0 or Veo 3.1. Your edited product shot becomes a short product film or video ad without leaving Avocado.
Can I edit images from Claude or another AI agent?
Yes. The Avocado MCP server gives Claude, Codex, Cursor or your own agent an edit_image tool that uses the same models and credits as the web editor. You describe the change in chat, the agent runs the edit, and the result lands in your Avocado library.