Product scene
Prompt
A matte white ceramic coffee mug on a light oak table, soft morning light from a window on the left, a few coffee beans scattered around, shallow depth of field, minimal Scandinavian style, square format, no text, no logos.Infinithink AI Image Generator
AI image generator for business visuals
Describe the picture you need and get product scenes, ad creatives, banners and social images, in the same workspace where you write the copy that goes with them.
What you can create with the AI image generator
Text-to-image AI works best when it has a clear job to do. Here are the jobs where it helps most.
-
Product scenes
Place a product type in a styled setting, such as a ceramic mug on an oak table in morning light. As an AI product image generator, it's good for mood shots, category banners and social posts. It creates new images from a description, so it can't reproduce your exact product. Use real photos when shoppers need to see exactly what they'll receive.
-
Ad creatives
Visual concepts and backgrounds for Meta and Google display ads. Write the headline with the AI content generator, then use the AI ad generator to test two or three visual directions before you brief a designer or pick a final image.
-
Banners and website headers
Wide images for a homepage hero, a seasonal promotion or a category page. Describe the mood, the colors and where you need empty space for text, and ask the AI banner generator to keep that area clear.
-
Posters and flyers
Backgrounds for event posters, menu boards and in-store signs. Image models can still get lettering wrong, so ask for space for the title and add the text in your design tool when every word must be exact.
-
Thumbnails and blog covers
Cover images for articles and YouTube videos. Thumbnails are seen small, so keep one simple subject and strong contrast. The AI thumbnail generator works best when you describe one clear focal point.
-
Social media images
Square and portrait visuals for Instagram, LinkedIn and Facebook that match the caption you just wrote. Generate a few versions and keep the one that fits your feed.
Image models available
Images are created with OpenAI's GPT Image model, on every plan.
The models in each plan are listed on the AI model comparison page.
How to write a prompt that works
A good prompt covers six things: subject, setting, style, light, format and what to leave out. Here are three prompts you can paste into the app as they are.
Ad visual
Prompt
Flat illustration for a Facebook ad for an accounting firm. A calm freelancer at a tidy desk with a laptop and neatly stacked receipts, indigo and cyan palette, plenty of empty space on the right for a headline, landscape format.Event poster
Prompt
Poster background for a summer jazz night in a town square, warm sunset colors, silhouettes of musicians on a small stage, string lights overhead, portrait format, top third left empty for the event title.Tips that make the biggest difference
-
Be specific about the subject and the setting. “A mug” gets you a stock photo. “A matte white mug on oak in morning light” gets you a direction.
-
Say where you need empty space for text.
-
Change one thing at a time when you refine.
-
Add “no text, no logos” unless you need them.
A simple workflow for campaign visuals
The same workflow works for a seasonal banner, a product launch post or a blog cover. It keeps the AI on what it does well (exploring visual ideas) and leaves the final layout to you.
-
Step 1, Start from the message
Write the headline and the offer first, so the image supports a clear idea.
-
Step 2, Generate three directions
Same subject, three different styles or settings. Pick the one that fits your brand.
-
Step 3, Refine one detail at a time
Light, background, framing. Small changes make it easier to see what works.
-
Step 4, Leave room for text
Ask for empty space where the headline and the button will go.
-
Step 5, Finish in your design tool
Add your logo, the exact text and your brand fonts, then export the sizes you need.
Images inside AI Chat
In the all-in-one AI assistant you can ask for an image in the middle of a conversation, then ask for changes in follow-up messages, such as “make the background darker” or “try a portrait version”. It's handy when the image idea comes out of a brainstorm.
Rights, labels and responsible use
Under our Terms, any rights we may have in the images you generate are assigned to you, so you can use them in your business. AI images may not be protected by copyright and similar images can be generated for other users, so don't rely on them as unique brand assets. Our Acceptable Use Policy forbids deepfakes of real people and images that infringe trademarks or other people's rights.
When an image could pass for a real photo of a real person, place or event, tell your audience it was generated with AI. Our AI Transparency Notice explains the rules in plain language.
Limits by plan
The Free plan includes a small number of images each month, enough to test the generator on your own ideas. Paid plans include more images every month, starting with Lite at €12 a month. The details are on the pricing page.
Pay yearly and get 2 months free.
FAQ
Is there a free AI image generator?
Yes. You can try it on the Free plan, which costs €0. You register with your email, and the plan includes a small monthly allowance of images. For regular use, paid plans start at €12 a month.
Can I use AI-generated images commercially?
Yes, within our Terms and Acceptable Use Policy. We assign you any rights we have in the images. Avoid real people's likeness, trademarks and copyrighted characters, and disclose AI generation when an image could mislead.
Which model creates the images?
OpenAI's GPT Image model, on every plan.
Can it design a logo?
It can explore logo ideas, styles and color directions. A finished logo needs clean vector files and a trademark check, so treat the images as concepts for a designer.
Why is the text in my image misspelled?
Image models still struggle with exact lettering. Ask for empty space instead, and add the words in your design tool.
How do I keep a consistent style across images?
Write one line that describes your style (palette, lighting, photo or illustration, level of detail) and reuse it word for word. Change only the subject. Keep that style line in a document in your workspace, so the whole team uses the same one.
Can I edit a photo I upload?
The generator creates new images from a text description. Editing your own photos isn't part of the image generator.
