Skip to main content

​
Quick Start

1

Navigate to Image Generator by Clicking on New Image

Go to the Image Generator tool in your Zoice dashboard by clicking on New Image.
New Image
2

Select AI Model

Choose an AI model (Imagen 4 Fast, Nano Banana 2, etc.) from the model selector.
Select Model
In the above image we have selected the Nano Banana 2 model.
3

Configure Settings

Set resolution, quality, and select text to image to get your idea into reality.
Configure Settings
In the above image we have slected the 1:1 aspect ratio and Ultra quality with text to image mode.
4

Write Your Prompt

Enter a detailed description of the image you want to create in the prompt box.
Write Prompt
5

Generate

Click generate and wait few seconds for your image to be generated.
Generate
6

Download

Save your generated image

​
Writing Effective Prompts

​
Prompt Structure

A well-structured prompt typically includes:
  1. Subject: What is the main focus?
  2. Style: What artistic style or aesthetic?
  3. Details: Colors, lighting, composition
  4. Quality: Resolution, detail level descriptors

​
Example Prompts

A red sports car on a mountain road at sunset

A sleek red Ferrari driving on a winding mountain road during golden hour, dramatic clouds in the sky, cinematic lighting, photorealistic, 4K quality

An abstract watercolor painting of a city skyline, vibrant blues and purples, flowing brushstrokes, modern art style, high contrast

A minimalist product photo of a smartwatch on a white marble surface, soft studio lighting, professional photography, clean composition, sharp focus

​
Generation Parameters

​
Resolution Options

  • Standard: 1024x1024 (fastest)
  • High: 1536x1536 (balanced)
  • Ultra: 2048x2048 (highest quality)

​
Aspect Ratios

  • Square: 1:1 (social media, profile pictures)
  • Portrait: 3:4 or 9:16 (vertical content)
  • Landscape: 16:9 or 4:3 (horizontal content)
  • Wide: 21:9 (cinematic)

​
Advanced Settings

Controls how closely the AI follows your prompt
  • Low (1-5): More creative, less literal
  • Medium (6-10): Balanced interpretation
  • High (11-15): Strict adherence to prompt
Use a specific seed number for reproducible results
  • Leave blank for random generation
  • Use same seed to recreate similar images
  • Useful for iterating on a concept
Generate multiple variations
  • 1-4 images per generation
  • Each variation is unique
  • Helps find the best result

​
Prompt Tips by Category

Key Elements:
  • Camera type and lens
  • Lighting conditions
  • Composition style
  • Depth of field
Example: “Professional portrait photo, 85mm lens, soft natural lighting, shallow depth of field, bokeh background”

​
Common Use Cases

Marketing Assets

Create social media graphics, ad visuals, and promotional images

Product Mockups

Generate product concepts and visualization

Blog Headers

Create custom header images for articles

Concept Art

Develop visual concepts for projects

​
Troubleshooting

  • Make your prompt more specific
  • Increase guidance scale
  • Try a different model
  • Regenerate with variations
  • Increase resolution setting
  • Add quality keywords to prompt
  • Use a higher-tier model
  • Specify “detailed” or “sharp focus”
  • Simplify your prompt
  • Remove contradictory descriptions
  • Be more explicit about what to exclude
  • Try negative prompts if available
Save prompts that work well! Build a library of effective prompts for different use cases.

​
Next Steps

Image to Image

Transform existing images

Explore Models

Learn about different AI models

Prompt Guide

Master prompt engineering