AI Image Editor — Edit Any Photo By Typing What You Want
Upload a photo, type an instruction in plain language — 'remove the person in the background', 'change the wall to sage green', 'make it look like golden hour', 'put the logo on the mug' — and the model does it. VdoBloom's Image Editor puts more than 30 image-edit models behind one dropdown, so you are not locked into whichever model your tool happens to license.
If you do not want to think about models, leave it on 'Auto — Smart Edit', which runs Seedream 4.5 first and falls back to WAN 2.7. If you do, the roster includes Nano Banana 2 and Nano Banana Pro, Seedream 4.5 Edit and Seedream 5.0 Pro, Flux Kontext Max, FLUX.2 [dev], Qwen Image 3.0 Edit, WAN 2.7 Image Pro, GPT Image 2, Kling IMAGE O1, Grok Imagine Pro, Riverflow 2.0 and Luma UNI-1 Max. Two specialist tools sit alongside them: FLUX Erase removes objects from a mask you paint, and FLUX Outpainting extends the canvas beyond the original frame.
What this tool does
- More than 30 image-edit models sit in one dropdown, including Nano Banana 2, Nano Banana Pro, Seedream 4.5 Edit, Seedream 5.0 Pro, Flux Kontext Max, FLUX.2 [dev], Qwen Image 3.0 Edit, WAN 2.7 Image Pro, GPT Image 2, Kling IMAGE O1, Grok Imagine Pro, Riverflow 2.0 and Luma UNI-1 Max.
- 'Auto — Smart Edit' picks for you: it runs Seedream 4.5 first and falls back to WAN 2.7 if that fails.
- FLUX Erase removes objects from a mask you paint directly on the image; FLUX Outpainting extends the canvas top, right, bottom or left beyond the original frame.
- Multi-reference editing limits by model: Grok Imagine Pro takes up to 20 reference images, Nano Banana 2 and Seedream 5 Lite up to 14, Kling IMAGE O1 and GPT Image 2 up to 10, WAN 2.7 up to 9, FLUX.2 [dev] up to 4.
- 4K output is available on Nano Banana Pro, GPT Image 2, WAN 2.7 Image, Seedream 5 Lite and Riverflow 2.0 Pro; output formats are PNG, JPG or JPEG depending on the model.
- An auto-translate toggle converts non-English edit instructions before they reach the model, and iPhone HEIC uploads are converted to a viewable preview automatically.
How it works
1.Upload the photo
Drop in a JPG, PNG or an iPhone HEIC file — HEIC is converted to a viewable preview automatically. Some models accept several reference images at once if you are combining elements.
2.Type the edit as an instruction
Write what should change, not what the picture is: 'remove the car on the left', 'replace the sky with a clear blue sky', 'change her jacket to red leather'. Naming what must stay the same — 'keep the face and pose identical' — measurably reduces drift.
3.Leave it on Auto, or pick a model
Auto — Smart Edit handles most edits. Reach for Flux Kontext Max or Seedream 5.0 Pro on precise local changes, Nano Banana 2 when you need many reference images, FLUX Erase to paint out an object, FLUX Outpainting to widen the frame.
4.Set aspect ratio and resolution, then generate
Resolution options depend on the model — 1K, 2K and 4K on Nano Banana Pro, GPT Image 2, WAN 2.7 and Riverflow 2.0 Pro. The credit cost updates on the button before you run it.
Frequently asked questions
What kinds of edits can it actually do?
Background replacement, object removal, object addition, colour and material changes, outfit changes, lighting and time-of-day changes, sky replacement, text and logo placement, style transfer, canvas extension and vectorisation. It is all driven by a typed instruction — there are no layers, brushes or selection tools to learn, except the mask painter used by FLUX Erase.
How do I remove an object or a person from a photo?
Select FLUX Erase, paint a mask over the object you want gone, and generate. The model fills the masked region from the surrounding image. For a quick removal without masking you can also just tell any edit model 'remove the person on the left, fill the background naturally'.
Can I expand a photo beyond its original edges?
Yes — FLUX Outpainting extends the canvas in any direction, with separate controls for top, right, bottom and left. It is the usual fix when a vertical phone photo needs to become a 16:9 banner without cropping the subject.
Which model should I choose?
Start with 'Auto — Smart Edit', which runs Seedream 4.5 and falls back to WAN 2.7. Switch deliberately when you have a reason: Seedream 5.0 Pro for coordinate-precise local changes, Flux Kontext Max for high-fidelity single-image edits, Nano Banana 2 or Grok Imagine Pro when you need to feed in many reference images, Nano Banana Pro or Riverflow 2.0 Pro when you need 4K out.
Can I write the instruction in Hindi or another language?
Yes. Turn on the auto-translate toggle and your instruction is translated before it reaches the model, which matters because most image-edit models are trained predominantly on English prompts.
Will it keep the person's face the same?
The identity-preserving models — Nano Banana Edit, Seedream 4.5 Edit — are built for this, and stating it in the instruction ('keep the face, hair and identity exactly as in the original') helps a lot. For dedicated identity work, the AI Headshots and AI Fashion Try-On tabs run these models at low edit strength with the preservation clause already written in.
Related tools
Photo to Prompt — Turn Any Image Into a Reusable AI Prompt
Turn any image into a structured AI prompt — pose, outfit, hair, lighting, camera and constraints, in a fixed section order. Send it straight to Edit.
AI Image Upscaler — Take Any Image to 2K or 4K
Upscale one image to 2K or 4K with Riverflow 2.0 Pro, or 2K with Riverflow 2.0 Fast. Face enhancement included. Runs in the browser, no install.
AI Product Photography — 24 Listing-Ready Shots From One Photo
One product photo becomes 24 shots: Amazon pure-white #FFFFFF, lifestyle, outdoor, studio, festive Indian, seasonal. Shape, colour and pack text locked.
AI Bikini Photo Generator — Swimwear Shoots From Text or One Photo
Generate swimwear photoshoots from text or one photo. Bikini, monokini and one-piece styling stays clothed — no nudity, adults 18+ only.
Ready to try it?
Open the AI Image Editor