DomoAI Text‑to‑Image Prompt Best Practice
How do I structure a good Text‑to‑Image prompt?
A strong prompt tells a complete story. According to an industry guide, include five elements: main subject, motion (if any), scene, environmental details and artistic style.
For example:
“A violinist playing under falling snow in a neon‑lit Tokyo alley, shot at dusk, watercolor style.”
What environmental details matter?
Think about time of day, weather, lighting, and perspective. Prompts like “early‑morning fog,” “golden‑hour backlighting,” or “low‑angle view looking up at skyscrapers” add depth and realism.
Do I need to mention artistic style?
Yes. Specify styles such as “oil painting,” “anime,” or “digital 3D render.” You can also reference artists or art movements to shape the aesthetic—for instance, “in the style of Studio Ghibli” or “cubist portrait.”
Or you can directly choose from DomoAI's preset image styles:
Can I keep prompts short and simple?
DomoAI will generate an image from simple prompts (e.g., “a serene landscape with mountains at sunset”). However, richer descriptions usually produce more refined results. Begin with a concise scene and iterate by adding details.
Does image quality depend on my prompt?
Yes. Clear, descriptive prompts with context and mood yield more accurate images. If your first result isn’t perfect, adjust the prompt—add or remove details and try again.
Nano Banana Pro Prompt Best Practice
What’s special about Nano Banana Pro prompts?
Nano Banana Pro is integrated with Google’s Gemini 3 and can produce native 4K images while blending up to 8 reference photos. Prompts should specify the subject, camera perspective, action, location and artistic.
How do I write prompts for complex scenes?
Break prompts into parts:
Subject – Describe the person or object in detail (“a Shiba Inu with metallic plating”).
Composition – Mention lens or viewpoint (“macro lens for texture,” “isometric view”).
Action – Add motion (“leaping across a rooftop gap”).
Location – Set the scene (“a neon‑lit Tokyo back alley”).
Style – Choose a medium (“vintage 1980s polaroid,” “oil painting”).
Also set the aspect ratio (e.g., 16:9 or 1:1) to avoid composition drift.
Can I blend multiple images?
Yes. Nano Banana Pro excels at multi‑image blending. When uploading several references, assign roles (“Image A for facial features, Image B for pose, Image C for background”) and consider starting with a few images before using all 8.
How do I add text in my images?
For legible text, enclose the words in double quotes and specify font, size and position.
For example:
”SUMMER SALE” in bold sans‑serif, 72 pt, centered at the top.
Keep text short (3–5 words); longer phrases may distort.
How can I keep characters consistent?
Use descriptive anchors (e.g., “woman with a left eyebrow scar”) in every prompt and maintain similar lighting and camera angles across the batch. When blending, specify each reference’s purpose and lock the random seed for consistent results.
What are popular use cases?
Nano Banana Pro supports:
Age progression or character timelines (e.g., photos of the same person across decades).
Product mockups with lifestyle backgrounds.
Sketch‑to‑reality transformations.
Memes and comics thanks to its precise text rendering.
Brand assets and marketing visuals, using brand guidelines as references.
Subject Swap
Ask LLM to provide the JSON format of this picture.
Using the JSON Format to change subject:
{
"scene_type": "first-person perspective",
"subject": {
"primary_object": "ice-cream",
"description": "a hand holding a ice-cream",
"position": "foreground, center"
},
"environment": {
"location": "urban street",
"setting": "narrow city road with buildings on both sides",
"time_of_day": "late afternoon",
"weather": "clear",
"lighting": "natural sunlight, warm tones"
},
"action": {
"activity": "riding a motorcycle",
"motion": "forward movement",
"camera_motion": "slight motion blur on background indicating speed"
},
"camera": {
"angle": "POV shot",
"lens_style": "wide-angle",
"depth_of_field": "shallow, subject in focus, background blurred"
},
"mood": {
"emotion": "romantic, hopeful, adventurous",
"atmosphere": "warm, cinematic"
},
"visual_style": {
"style": "cinematic photography",
"color_palette": "warm highlights, earthy tones, vivid red accents",
"realism": "photorealistic"
},
"details": {
"vehicles": "cars parked and driving along the street",
"architecture": "residential buildings with balconies",
"props": [
"motorcycle handlebars",
"speedometer",
"bouquet of roses"
]
}
}
Are there any limitations?
Nano Banana Pro struggles with long or curved text, maintaining consistency for more than five people, and complex spatial relationships. When you notice issues, reduce prompt complexity or refine one element at a time.