Skip to main content
Image Prompt lets you combine a reference image with a text prompt to guide the AI toward a specific visual direction. Rather than relying on words alone, you provide a visual anchor—a photo, sketch, or previous generation—that shapes the structure, subject, or feel of the output. Your text prompt then steers the AI’s creative choices on top of that foundation. Think of it as a conversation: the image sets the scene, and your text provides the direction.

​
How to use Image Prompt

1

Open the settings panel and enable Image Prompt

In Image mode, open the settings panel. Click Image Prompt to enable it. An upload area with a + button will appear.
2

Upload your reference image

Click the + button or drag your reference image directly into the upload area. You can use:
  • A photo you’ve taken
  • A sketch or illustration
  • A previously generated image
  • Any image that captures the visual element you want the AI to reference
Once uploaded, the image is set as an active visual reference for the next generation.
3

Write your text prompt

In the main prompt box, describe what you want to create. Your text prompt works together with the image reference—the AI blends guidance from both. Be specific about what aspects of your vision to emphasize.Example combinations:
  • Reference image of a misty forest + prompt: "A lone cabin with warm glowing windows, winter evening, oil painting style"
  • Reference image of a sports car + prompt: "Same car on a coastal highway at sunset, cinematic photography, motion blur"
  • Reference image of a character sketch + prompt: "Full color illustration, fantasy armor, dramatic lighting"
4

Select your model and settings

Image Prompt is supported by all available image generation models. Choose the model that best suits your output goal, then set your preferred aspect ratio and number of images.
Models like Flux Dev and Seedream v3 tend to interpret visual references expressively, while models like Imagen 3 and Flux Ultra will produce more photorealistic outputs grounded in the reference.
5

Iterate as needed

After reviewing the initial output, you can:
  • Adjust your text prompt to shift emphasis
  • Swap in a different reference image
  • Try a different model to change how the reference is interpreted
  • Generate additional batches to explore variations
6

Click Create

Click Create to generate. The AI will synthesize your reference image and text prompt into a new image.

​
Supported models

Image Prompt works with all image generation models available on ImagineArt, including:
  • Flux Dev, Flux 2 Pro, Flux 2 Max, Flux 1.1 Ultra
  • ImagineArt 1.0, 1.5, 1.5 Pro
  • Nano Banana, Nano Banana 2, Nano Banana Pro
  • ChatGPT Image
  • Imagen 3, Imagen 4
  • Ideogram v3, Ideogram Character
  • Minimax Image
  • Seedream v3, v4, v4.5, v5 Lite
  • Dreamina 3.1
  • Qwen Image
  • Recraft v4, Recraft v4 Pro
  • xAI Grok Imagine, Z Image Turbo
The model you choose affects how strongly the reference image influences the output. Experiment with different models to find the balance between reference fidelity and creative freedom that works best for your project.