Step 1
Write a clear prompt
Describe the subject, setting, style, lighting, and details you want. You can also explain what should stay unchanged when you are editing an existing image.
AI Image Generator
Create new images from a prompt or guide the result with up to five reference images. Choose the model, dimensions, resolution, and number of outputs before you generate.
Results
Explore these examples to get inspired. Your generated images will appear here.
💡 Pro tip: Hover overDrag on the examples below to see the before/after transformation effect
"Camera is closer to the cat who's now looking at the camera"
BeforeAfter
"She's now holding an umbrella and smiling"
BeforeAfter
"Give the ball stick man cartoon arms and legs and a simplistic face and make him go on a happy stroll"
BeforeAfter
"Replace "Joy" by "BFL""
BeforeAfter
How the AI Image Generator works
Create a new image from text, guide the result with reference images, or use both. The controls stay in one form, so you can adjust the model and output settings before spending credits.
Step 2
Add reference images when useful
Upload up to five reference images to guide the subject, composition, product details, or visual style. References are optional for text-only generation.
Step 3
Choose the output settings
Select an image model, aspect ratio, and resolution that fit the job. The form shows the credit cost before you generate.
Step 4
Generate and compare results
Create one to four images in a batch. Review the results, adjust the prompt or settings, and run another version when you need a different direction.
AI Image Generator features you can control
The available controls depend on the model you select. The page shows compatible image inputs, dimensions, resolutions, and pricing as you configure the result.
Text and reference image input
Start from a written idea or use existing images as visual guidance. This works for new compositions as well as controlled changes to a source image.
Several image models in one workspace
Choose from Kontext, Nano Banana, GPT Image, Qwen Image 3, Flux 2, Seedream 5 Pro, Wan 2.7 Image, and Grok Imagine without changing tools. Each model lists its supported inputs, sizes, and resolution options.
Flexible dimensions and resolution
Use automatic sizing, match a reference image when supported, or select a fixed aspect ratio. Available resolution choices depend on the selected model.
Small batches for faster comparison
Run up to four generations from the same setup. Some models return several variations per generation, making it easier to compare results before refining the prompt.
AI Image Generator use cases
Use the generator for quick visual drafts or finished creative assets. Reference images help when a result needs to follow an existing subject, product, or style.
Marketing and social content
Create campaign concepts, social graphics, ad visuals, and alternate compositions for different placements.
Product and ecommerce images
Build product scenes, test backgrounds, or create lifestyle concepts while using reference images to guide important details.
Concept art and visual exploration
Try character ideas, environments, color treatments, and art directions before committing to a final design.
Photo changes and creative variations
Change a background, add or remove an object, explore a new style, or produce several interpretations of the same image.
AI Image Generator Frequently Asked Questions
Learn how to choose a model, resolution, and workflow for your next image.
Which image model should I choose?
Kontext Pro is a strong default for reliable generation and editing, while Kontext Max prioritizes maximum quality. Nano Banana is an affordable everyday option, Nano Banana Pro targets maximum fidelity, Nano Banana 2 balances speed and quality, and Nano Banana 2 Lite is best for fast drafts. GPT Image 2 is a good choice for precise instructions and polished visual content. Qwen Image 3 handles rich layouts and multilingual text, Flux 2 is suited to photoreal detail, Seedream 5 Pro supports controlled multi-reference work, Wan 2.7 Image offers flexible generation and editing, and Grok Imagine quickly returns several prompt variations.
How do model and resolution affect credits?
Credits are charged per generation. Most models return one image, while Grok Imagine returns several variations from a text prompt. Some prices are fixed and others vary by resolution. The form always shows the output count and full total before you submit.
Can I use the generated images for commercial purposes?
Our plans support commercial creative workflows, including marketing materials, product visuals, and client projects. Usage remains subject to our Terms of Service and the rules that apply to the selected model.
What file formats are supported for reference images?
You can upload up to 5 PNG, JPG, JPEG, GIF, or WebP images. Each image can be up to 10 MB.
How do I write effective prompts for image editing?
For best results, write prompts in English and be specific with colors and details. When editing existing images, mention what to preserve from the original (e.g., "change hair color to blonde while keeping the same facial features").
Use descriptive adjectives like "soft watercolor" or "vintage style" and specify lighting and mood when relevant. For text replacement, use quotation marks around the exact text you want to replace.
What industries and use cases work best with these models?
The available models work well across many industries:
- Marketing & Advertising: Rapid campaign asset creation and brand content updates
- E-commerce: Product photography enhancement and lifestyle imagery
- Entertainment & Media: Character development and concept art generation
- Education: Historical recreations and visual storytelling
How long does it take to generate images?
Generation times typically range from 30 seconds to 2 minutes depending on the complexity of your prompt, the model selected, and current system load. You can run up to four generations in one batch to maximize efficiency.
Do you offer an API for developers?
Yes! We provide developer-friendly API integration with comprehensive documentation and support for multiple programming languages. This makes it easy to integrate AI image generation into your own applications and workflows.