Image Generator

Create images from natural-language text prompts in seconds. Generate multiple variations, explore creative ideas, and bring your vision to life with AI-powered image generation.

What you can do:

Generate up to 4 images from a single text prompt
Choose from multiple AI models or let the system auto-select
Control style, camera angle, and color palette
Use templates for quick generation or customize every detail
Create marketing visuals, product imagery, and design assets

How it works:

The Image Generator uses AI models to interpret your text descriptions and create corresponding images. Each generation is unique, running the same prompt may produce different creative variations, giving you multiple options to explore.

Quick Start

Select a model, or choose Auto to let the system decide.
Enter a text prompt describing the image you want to generate.
(Optional) Enhance the prompt for improved clarity.
Use references and controls to guide the visual output.
Generate images.

Each generation can produce up to four images using the same configuration.

Use Cases

The Image Generator is suited for a wide range of creative and exploratory workflows where images are generated from text descriptions and refined through visual guidance.

Common use cases include:

Rapid ideation and concept exploration: Generate quick visual concepts from simple prompts to explore ideas, compositions, or moods without committing to a specific style upfront.
This is useful during early-stage brainstorming, creative exploration, or when evaluating multiple directions for a visual idea.
Marketing and content visuals: Create images for blogs, presentations, landing pages, or social media by combining descriptive prompts with style, camera, and color references.
Generating multiple images at once helps compare variations and select the most suitable output.
Product and lifestyle imagery: Generate clean, stylized product visuals or contextual lifestyle scenes by describing the product and environment in the prompt and refining the output using references.
This is useful for mockups, previews, or conceptual product shots.
Illustrations and design assets: Create illustrations, design-oriented visuals, or stylized graphics by selecting illustration or design styles and pairing them with concise prompts.
This workflow works well for UI concepts, posters, thumbnails, or creative assets that do not require photographic realism.
Visual experimentation and variation: Generate multiple images from the same prompt to explore different interpretations, styles, or compositions.
This is useful when variation and creative diversity are desired rather than a single fixed outcome.

Custom Generation

Custom generation allows you to create images by manually configuring prompts, references, and controls. This mode provides full control over how images are generated and is suited for custom or exploratory workflows.

Model Selection

Image generation is performed using one of the available models. Each model is optimized for different styles and generation characteristics.

Available options include:

Nano Banana
Nano Banana Pro
Seedream 4.5
Imagen 4
Imagen 4 Fast
Imagen 4 Ultra
Auto

Selecting Auto allows the system to automatically choose a model based on the prompt and configuration. This option is useful when users are unsure which model best fits their use case.

Prompt

The prompt defines what should appear in the generated image. It can describe the subject, environment, mood, style, composition, or any other relevant visual detail.

Clear and specific prompts generally produce more predictable results, while vague prompts encourage broader variation.

Prompt enhancement

You can optionally enhance your prompt before generation. Prompt enhancement restructures and expands the input text to improve descriptive clarity while preserving the original intent.

Prompt enhancement helps reduce ambiguity and can improve results by making your description more precise.

Prompt guidance

When writing prompts:

Start with the main subject
Add context or environment
Specify style, lighting, or mood if important
Use references and controls for stylistic guidance instead of overloading the prompt

tip

Start with a clear subject and add style or mood later. Overloading the prompt with conflicting instructions often leads to inconsistent results.

Example Prompts

Simple subject:

A modern workspace with a wooden desk, laptop, and indoor plants

Descriptive scene:

A cozy reading corner by a large window, warm sunlight, soft shadows, minimal interior design

Stylized illustration:

A futuristic city skyline at night, neon lights, cyberpunk aesthetic, digital illustration

Product focused image:

A white ceramic coffee mug on a marble table, studio lighting, clean background, professional photography

These examples illustrate how increasing descriptive detail influences the generated output.

note

Image generation is unique each time. Using the same prompt and settings may produce different creative variations.

References

References provide additional control over the visual characteristics of the generated image. These options guide style, composition, and color treatment independently of the text prompt.

Style

Determines the overall visual style applied to the image.

Available categories include:

Photography: Natural, editorial, landscape
Illustration: Cartoon, pixel art, oil painting
Design: Inkprint, isometric, neonwave, minimalist

Selecting Random applies a randomly chosen style to the output.

Camera

Controls camera perspective or framing.

Examples include:

Drone
360
Portrait
Close-up
First-person
Aerial

Selecting Random applies a randomly chosen camera style.

Palette

Controls the dominant color palette of the generated image.

Available options include:

Cold
Neon
Pastel
Vibrant
Warm

Color palettes influence the overall mood and tone of the image.

Upload Reference

Users can upload images to act as visual references. Reference images can influence style, color treatment, or composition.

Uploaded references supplement the prompt and reference settings rather than replacing them.

note

Reference images guide style and color while allowing creative interpretation in the generated output.

Aspect Ratio

Select the aspect ratio for your generated images.

Available options:

1:1: Square format
16:9: Widescreen landscape
9:16: Vertical portrait

note

All images generated in a single request use the same prompt, references, and aspect ratio.

Templates

Templates provide a faster way to generate images using predefined settings. Each template includes a curated combination of prompt structure, style, camera settings, and reference inputs.

Unlike Personal generation where users configure every parameter from scratch, templates are designed to reduce setup time and decision overhead.

How Templates Work

When a template is selected, the Image Generator automatically applies a predefined configuration that may include:

A structured prompt
Style and camera settings
Color palette choices
Reference images

Users can generate images immediately using the template defaults or modify the prompt and settings before generation.

note

Templates act as starting points, not fixed workflows.

When to Use Templates

Templates are useful when:

You want to generate images quickly without configuring multiple options
You are exploring a common or repeatable visual style
You prefer guided generation over manual setup
You are new to image generation and want sensible defaults

They are especially effective for recurring use cases such as marketing visuals, lifestyle imagery, or design assets.

Custom Generation vs Templates

Custom Generation:

Full control over prompt and parameters
Best for custom or experimental generation
Requires manual configuration

Templates:

Predefined parameters and references
Faster setup and lower decision effort
Ideal for quick generation and consistent outputs

Both approaches use the same underlying generation system and credit model.

Customizing Templates

Templates are not locked. After selecting a template, users can:

Edit the prompt
Adjust references and controls
Change aspect ratio or number of images

Any changes override the template defaults for that generation.

Quick Reference

Specification	Details
Credits per image	1.5 - 11 (varies by model)
Max images per generation	4
Available models	Nano Banana, Nano Banana Pro, Seedream 4.5, Imagen 4, Imagen 4 Fast, Imagen 4 Ultra, Auto
Aspect ratios	1:1, 16:9, 9:16
Style categories	Photography, Illustration, Design
Camera options	Drone, 360, Portrait, Close-up, First-person, Aerial, Random
Color palettes	Cold, Neon, Pastel, Vibrant, Warm
Generation modes	Custom Generation and Templates

Credit Consumption

Credit consumption varies by model:

Nano Banana: 3 credits per image
Nano Banana Pro: 11 credits per image
Seedream 4.5: 3 credits per image
Imagen 4: 3 credits per image
Imagen 4 Fast: 1.5 credits per image
Imagen 4 Ultra: 5 credits per image

For example, generating four images with Nano Banana consumes 12 credits, while generating four images with Imagen 4 Fast consumes 6 credits.