# AI Photo Editor > Upload your photos, describe the edit, and keep the subject. Change hair, lighting, objects, or wardrobe from a prompt, then download the result on MixVio. - [Canonical AI Photo Editor page](https://mixvio.ai/tools/ai-photo-editor) ## Tool summary - **Media:** image - **Input:** Owned photos plus a keep-and-change instruction - **Output:** Private edited image as a workspace PNG - **Credits:** Quoted from model, quality, and resolution - **Limit:** Reference count follows the selected edit model ## How to use 1. **Upload your photos:** Drop in portraits, product images, or rooms you own or are authorized to edit. One image starts a run; the selected model sets how many references you can add. 2. **Describe the edit:** Type what should change and what must stay—face, product, or room. Example: keep the face, change only the dark wavy hair to a platinum bob. 3. **Generate and download:** Confirm the credit quote, generate, inspect the result, then download or run another pass on the same image. ## Common uses - Portrait cleanup and hair changes - Product and room one-detail fixes - Lighting and object-state edits ## FAQ ### What is an AI photo editor? AI Photo Editor starts from photos you already own and keeps the subject while you change one thing you ask—cleanup, lighting, hair, clothes, or a single object. It is not a layers-and-sliders editor, not text-to-image from a blank canvas, and not Image to Video. MixVio quotes credits before you run. ### How do I edit a photo with AI? Three steps in the generator on this page. 1) Upload your photos: drop in portraits, product images, or rooms you own or are authorized to edit. Generate stays disabled until at least one photo is attached. 2) Describe the edit: type what should change and what must stay—face, product, or room. Example: keep the face, change only the dark wavy hair to a platinum bob. 3) Generate and download: confirm the credit quote, generate, inspect the result, then download or run another pass. ### Is MixVio AI Photo Editor free? An account is required. New accounts receive 40 welcome credits valid for 7 days. AI Photo Editor is available on the Free plan when your balance covers the quote. GPT Image 2.5 starts at 2 credits for Low at 1K. Free image-edit delivery is watermarked and capped at 1K; paid plans unlock higher quality and resolution. Failed runs cost no credits. ### How do credits work? Credits are quoted from the selected model, quality, and resolution before you generate. GPT Image 2.5 starts at 2 credits for Low at 1K. The default 2K Medium pass is 3 credits; other models on this page quote separately. Failed or cancelled tasks release the reservation. ### What kinds of edits can I make? Retouch a portrait, change hair or clothes, remove one object, relight a room or product, clean a desk, or update a label. Name the face, product, or room that must stay. This page does not cut a transparent background, upscale pixels, or restyle the whole scene—use Background Remover, Image Upscaler, or Image to Image for those jobs. ### What kinds of prompts work best? Name what must stay and the one change in the same instruction. Keep identity, product, or room layout explicit. Example: keep the same woman, face, and navy cardigan; change only the dark wavy hair to a platinum bob. Vague “make it better” prompts leave the model guessing. ### Do I need a brush or mask? No. You do not select a region first. Write what should change and what must stay. Unrequested details can still shift, so inspect identity, labels, logos, and fine typography before delivery. This page does not expose a brush or mask canvas. ### What if I am not satisfied with the result? Adjust the keep-and-change instruction, switch edit model, or replace the source photo, then generate again. Each new run shows its credit quote before submit. Failed runs cost no credits. ### Which models can I use on AI Photo Editor? The public tool defaults to GPT Image 2.5 for keep-and-change edits and lets you switch to GPT Image 2, Seedream 5.0 Pro, Seedream 5.0 Lite, Nano Banana Pro, Nano Banana 2, and Nano Banana 2 Lite. Pick by subject lock, prompt control, layout, or speed. Each route stays on the same image-edit contract. Open a model page when you need resolution, reference limits, or credit details. ### What photos can I upload? Upload owned JPEG, PNG, or WebP photos. Each run needs at least one source photo. The selected model sets the count and size limits—GPT Image 2.5 accepts up to 16 photos; other edit routes on this page may accept fewer. This is not a folder batch: edit the attached photos in one prompt, then start another run for the next set. MixVio returns a private edited PNG to your workspace; inspect it before you download. ### Can I use AI Photo Editor outputs commercially? Paid-plan outputs may be used commercially, subject to the Terms of Service and the current provider terms. You must have rights to every source photo, face, brand, and prompt. Free-plan outputs are for evaluation and are not for commercial publishing. ### Are uploads and results private? Yes. Uploads and outputs remain private workspace assets. Signed URLs, asset IDs, and private prompts are not added to public page URLs or analytics. ### Should I use AI Photo Editor or Image to Image? Use AI Photo Editor when the photo is already right and you need one local fix—cleanup, relight, retouch, or a single object. Use Image to Image when you want a broader restyle, rebuild, campaign restage, or sketch-to-image pass. Background Remover returns a transparent PNG cutout. Image Upscaler sharpens and enlarges a photo after the creative content is already set. [Open Image to Image →](https://mixvio.ai/tools/image-to-image)