Nodaro Docs
DocumentationNode ReferenceModelsAI Agents (MCP)DevelopersSelf-hostingResearch
Image

Modify Image

Change an existing image with a text instruction. Restyle it, fix one detail or repaint a masked area with more than 20 image-editing models.

The Modify Image node changes an existing image according to a text instruction. You connect a picture, choose one of more than 20 image-editing models, and describe the change: a new style, a new background or one fixed detail. The node returns a new image and leaves the original untouched.

When to use it

  • Restyle a photo or a render, for example into watercolor, anime or a film look.
  • Make a targeted change, such as "add a hat to the person" or "make the sky stormy".
  • Repaint one area only. Paint a mask, or wire one from Generate Mask or Paint Mask.
  • Give an image a new shape and let the model fill the new space.
  • Apply the same edit to many images. Wire a List of images into the Image input, and the node runs once for each image.

To enlarge an image without changing it, use Upscale Image. To cut out the subject, use Remove Background. Neither needs a prompt.

Quick start

Add the node

Press Tab on the canvas and choose Image › Edit & Retouch › Modify Image.

Connect the image

Drag from an image node, such as Upload Image or Generate Image, to the Image input.

Choose a model and describe the change

Open the settings panel and choose a model under Provider. Write the change in Transformation Prompt. With Nano Banana Edit, the same field is called Edit Instructions.

Run it

Click Run on the node. The new image appears on the node. Every run adds a version, and earlier versions stay on the node.

imagecinematographyimageUpload ImageStreet photoLightingBlue hourModify ImageNano Banana EditUpscale ImageTopaz Upscale, 2x
A photo is edited by instruction with a Lighting picker, then upscaled for the final file.

Inputs

InputAcceptsWhat it does
ImageImage nodes, such as Upload Image, Generate Image and Extract Frame, or a List of imagesThe picture to change. Required.
MaskImage nodes, such as Generate Mask, Paint Mask and Image OverlayThe area to repaint: white areas change, black areas stay. This input appears only when the selected model supports masks, such as Ideogram Edit.
CinematographyLook pickers, such as Style, Lighting, Lens, Color / Look and MoodEach picker adds its choice to the end of the prompt.

The output, Image, is the new picture. It can feed any number of nodes at once.

Settings

The settings panel shows only the settings that the selected model supports. When you switch models, a value the new model does not accept is reset.

SettingWhat it does
ProviderThe editing model. The default is Nano Banana.
Transformation PromptWhat to change. With Nano Banana Edit, the field is called Edit Instructions.
StyleNo style, one of 16 style presets such as Photorealistic, Anime or Watercolor, or Custom... for your own words. The style is added to the prompt. When a Style picker is connected, the picker is used instead.
Negative PromptWhat to avoid. Some models receive it as a real negative prompt; for the others, it is added to the prompt.
AssetsCharacters, locations and objects from your library, added with Add from Library or Create new. Their pictures and descriptions help the model keep them consistent.
Inpainting MaskPaint the area to change over the connected image. Shown only on models that support masks. See Change one area with a mask.
Aspect RatioThe shape of the result. The choices depend on the model.
ResolutionThe output size, on models that have it, such as 1K, 2K or 4K. The two Flux 2 (Safety Tolerance) models set it in megapixels, from 0.5 MP to 4 MP.
QualityA quality tier on models that have one: medium or high on GPT Image, basic or high on Seedream.
StrengthHow far the result may move away from the original. Lower stays closer to the original; higher is more creative.
Guidance ScaleHow strictly the model follows the prompt. Higher is stricter.
Rendering SpeedOn Ideogram models: Turbo is faster and cheaper, Balanced is the default, and Quality is the best and costs more.
SeedA fixed number that makes a run repeatable. Leave it empty for a random seed.
Reference ImageAn optional extra picture that guides the style, added with Upload Reference. Shown on models that accept references.
Pre & post textText that is always added before and after the prompt. It is hidden from people who use your workflow as an app. See Prompt pre and post text.

Models

Modify Image can run every model in the table. Click a model for its settings, credit prices and prompt tips.

ModelMakerModesCreditsDetails
Nano Banana 2GoogleText to image, Image to imagefrom 20Newer Nano Banana with native resolution control (1K/2K/4K) and Google Search context.
Nano Banana 2 LiteGoogleText to image, Image to image10Lightweight Nano Banana 2 (Gemini 3.1 Flash-Lite) — fast, low-cost 1K generation and editing.
Nano Banana ProGoogleText to image, Image to imagefrom 45Top-tier Nano Banana — best for text rendering, diagrams, and complex compositions.
Nano Banana EditGoogleImage editing10Image-to-image edits via Google's Nano Banana family. Good general-purpose editor.
Flux 2 Flex (I2I)Black Forest LabsImage to image60Image-to-image with Flux Flex. Honors source image structure while applying prompt.
Flux 2 Pro (I2I)Black Forest LabsImage to imagefrom 13Image-to-image with Flux Pro. Cheaper than Flex variant, good general edits.
Flux Kontext ProBlack Forest LabsText to image, Image editing13Context-aware editing and style transfer. Strong at preserving subject identity through edits.
Flux Kontext MaxBlack Forest LabsText to image, Image editing25Premium Kontext — highest fidelity context-aware edits.
Kontext Multi (Open)Black Forest LabsImage to image, Image editing30Multi-image Flux Kontext Pro — up to 4 input images.
Flux 2 Pro (Safety Tolerance)Black Forest LabsText to image, Image to imagefrom 12Accepts up to 4 reference images.
Flux 2 Max (Safety Tolerance)Black Forest LabsText to image, Image to imagefrom 18BFL Flux 2 Max — even larger sibling of Pro, safety_tolerance=5, up to 8 reference images. Variable pricing by MP and ref count.
GPT Image 2 (I2I)OpenAIImage to imagefrom 15Image-to-image with GPT Image 2.
GPT Image 2.5 Flare (I2I)OpenAIImage to imagefrom 15Fast GPT Image 2.5 edits (up to 16 source images) - the default when you are iterating rather than finishing.
GPT Image 2.5 Sunburst (I2I)OpenAIImage to imagefrom 15Precision GPT Image 2.5 edits (up to 16 source images) - the most controlled edit in the GPT family, at the cost of a longer run.
Ideogram EditIdeogramImage editingfrom 30Inpainting / mask-based editing with Ideogram. Pair with a mask URL.
Ideogram RemixIdeogramImage to imagefrom 30Ideogram remix — character-aware restyling driven by reference images.
Seedream 5 Lite (I2I)BytedanceImage to imagefrom 14Image-to-image with Seedream 5 Lite.
Seedream 5 Pro (I2I)BytedanceImage to imagefrom 19Image-to-image with Seedream 5 Pro — multi-reference instruction edits. Basic = 1K, high = 2K.
Qwen (I2I)AlibabaImage to image10Image-to-image with Qwen.
Qwen EditAlibabaImage editing13Qwen image edit endpoint with native negative prompt.
Grok Imagine (I2I)xAIImage to image10Image-to-image with Grok.

The Provider list also offers Nano Banana, the default, plus GPT Image and Seedream Edit. These three models have no model page.

Which model to choose

For a comparison of every image model, read Choosing a model.

Change one area with a mask

A mask tells the model which pixels it may repaint. White areas change; black areas stay as they are.

Choose a model that supports masks

Choose Ideogram Edit under Provider. The Mask input and the Inpainting Mask section appear.

Paint the mask

Click Paint Mask in the Inpainting Mask section. Paint over the area to change, then click Save mask. The section now shows the mask, with Edit Mask and Clear Mask.

Describe the change and run

Write what the painted area should become, for example a red umbrella, and run the node.

Instead of painting in the panel, you can wire a mask into the Mask input. Generate Mask finds a subject from a description such as "the red car". Paint Mask keeps a hand-painted mask as its own node, which several nodes can share. When you reopen the painter on a saved mask, it starts from that mask, so you can touch it up instead of starting again.

Change the aspect ratio

To extend an image to a new shape, choose Nano Banana Edit, set the target Aspect Ratio, and use an instruction such as:

Prompt
extend the scene naturally into the new space, matching light and perspective

Start from a preset

Modify Image shares a set of transform presets with Generate Image. Open the preset menu on the node and choose one:

  • Stylized Subject — change the style of the person but not the world around them, or the other way round. For example, a cartoon person in a real photo, or a real person in a cartoon world.
  • Edits — remove or replace the background, colorize, restore an old photo, relight or restyle. Two Doodle Overlay presets draw marker doodles over the photo and leave the pixels underneath unchanged.

These presets keep their instruction in the pre and post text, so the prompt field stays free for your own words. See Presets.

Credits

The cost depends on the model and on its quality, resolution or rendering-speed setting. The Run button shows the exact price before you run. Flux 2 Max (Safety Tolerance) is priced by output size and by the number of reference images. Each model page lists its prices.

Tips

  • Describe the change, not the scene. "Make the sky purple" works better than "a landscape with a purple sky".
  • Start with a low Strength. On models that have it, begin around 0.3 to 0.5 to keep the original structure, then raise it for bigger changes.
  • Restyle a character with Ideogram Remix. It keeps the identity while the style changes.
  • Keep brand assets as files. A model redraws a logo slightly wrong. Edit the picture first, then place the real logo with Image Overlay.
  • Chain simple steps. For example, remove the background, then upscale the cut-out, instead of asking one model for both.

Frequently asked questions

Last updated on

On this page