Skip to main content
Recraft Studio supports targeted image editing using integrated external models Google Nano Banana, OpenAI GPT‑4o, Seedream, Flux Kontext, and Qwen-Image that respond to natural language prompts and optional reference images. These models are built directly into the Recraft interface, so you can make precise edits without switching platforms or starting from scratch. Natural language editing allows you to modify specific aspects of an image (like clothing, background, or facial expression) while preserving the overall structure and style. Only the elements described in your prompt are changed; everything else remains intact.

​
Use cases

  • Make small adjustments, such as changing clothing, background elements, or facial features
  • Refine visual details without regenerating the entire image
  • Maintain stylistic or character consistency across multiple images
  • Add, remove, or modify specific objects within a scene
  • Adjust visual elements like lighting, color palette, or seasonal context
  • Update product features or character traits while keeping composition intact
  • Apply a consistent style or mood using reference images

​
How to edit an image with natural language prompts

  1. Open a project and upload or generate the image you want to modify.
  2. Select the image on the canvas.
  3. In the prompt panel, select a model:
    • Nano Banana (normal, Pro, Pro 4K)
    • Seedream V4
    • GPT‑4o (Low, Medium, or High quality)
    • Flux Kontext (Pro or Max quality)
    • Qwen-Image
  4. Enter a prompt describing the change (e.g., “make the jacket red” or “change background to nighttime”).
  5. (Optional) Click the paper clip icon within the prompt panel to attach reference images:
    • Up to 9 images for GPT‑4o, Nano Banana, and Seedream
    • 1 image for Flux Kontext and Qwen-Image Reference images allow the model to incorporate visual elements like style, character, pose, or layout into the output. In your prompt, describe what you want to borrow from each image and how you’d like them composed.
  6. Click the generate button to apply the edit.
  7. You can repeat the process as needed. Each new edit builds on the previous result without requiring a full regeneration.

​
Supported models

Currently, the reference image feature is supported only by select external models. You can manually choose the model, or GPT‑4o Medium will be used by default when a reference image is attached. Recraft standard image generation is included in the table below for reference.

​
Notes on external models

Each external model has its own supported aspect ratios. For example, GPT‑4o supports only 1:1, 3:2, and 2:3 formats. If your original image uses an unsupported aspect ratio (e.g., 16:9), the output will default to the closest supported format. The credit cost for using external models depends on each model’s API pricing.