Aivon Vision is an AI-powered creative platform that lets you generate stunning images, transform existing photos, and remove backgrounds - all in seconds. This documentation will help you get the most out of every credit and every generation.
Aivon Vision currently offers two core tools. Each uses cutting-edge AI models and costs 1 credit per successful generation.
Describe anything in plain language and watch it come to life. Supports multiple aspect ratios and 10 art styles.
Instantly remove backgrounds from any photo with AI precision. Get a clean transparent PNG ready for use in designs.
Click on each tool below for a step-by-step walkthrough.
Click Generate in the sidebar. Make sure the Text to Image tab is selected at the top.
Describe the image you want in the prompt box. Be specific - include subject, setting, lighting, mood, and style. Up to 1000 characters.
Pick an art style from the dropdown (see Supported Styles section). This is optional, you can describe your styling or atmosphere in the prompt too.
Click Generate (1 credit) or press Ctrl+Enter. Your image will appear in 5-20 seconds.
Click Download to save the PNG. All generations are saved to your History automatically.
Click Remove BG in the sidebar.
Click Choose Image or drag & drop. PNG, JPG, and WEBP are supported up to 10MB.
Click Remove Background. The AI will detect and remove the background in seconds.
Download the result as a transparent PNG ready to use in any design tool.
Choose a style from the dropdown on the Generate page to guide the AI's visual output.
The quality of your prompt directly determines the quality of your output. Here are examples of weak vs strong prompts:
| ❌ Weak prompt | Why it fails | ✅ Strong prompt |
|---|---|---|
| a cat | Too vague - no context, lighting, or style | a fluffy orange tabby cat sitting on a windowsill, golden hour sunlight, photorealistic, sharp focus |
| city at night | No mood, perspective, or detail | aerial view of Tokyo at night, neon reflections on wet streets, cyberpunk aesthetic, dramatic lighting |
| portrait of woman | No style, setting, or character detail | portrait of a young woman with braided hair, soft studio lighting, minimalist background, oil painting style |
| fantasy scene | No subject, scale, or atmosphere | ancient dragon perched on a mountain peak, storm clouds, epic fantasy art, cinematic volumetric lighting |
Follow these practices to produce great results consistently.
The more detail you provide, the better the AI understands your intent. Include subject, environment, lighting, mood, and style in every prompt.
Instead of describing the style in your prompt, use the built-in style dropdown. This gives the model a clearer, more structured signal and produces more consistent results.
Use 1:1 for profile pictures and social posts, 16:9 for desktop wallpapers and YouTube thumbnails, and 9:16 for mobile wallpapers and Instagram stories.
When you get a result you like, use the "Generate Again" button with a slightly modified prompt to explore variations rather than starting from scratch.
More than 150-200 words rarely improves results and can confuse the model. Focus on the most important elements and leave the rest to the AI.
Each generation costs 1 credit. Here's how to make every credit count.
| Tool | Cost | Notes |
|---|---|---|
| Text to Image | 1 credit | Charged only on successful generation |
| Remove Background | 1 credit | Charged only on successful processing |
Spend 30 seconds refining your prompt mentally before clicking Generate. A well-crafted prompt on the first attempt saves credits versus generating 3-4 times to get it right.
Before generating something new, check your History page - you may have already created something similar. All generations are saved automatically.
All packs are priced at €0.05 per credit. Credits never expire, so buying a larger pack upfront costs the same per credit but saves you the hassle of topping up frequently.
Check our FAQ for quick answers to common questions, or reach out to our support team.