About Image to prompt
You like an image and want something close to it, but cannot put its lighting, angle and style into words a generator understands. Or you have a product photo and need a persuasive listing for your shop. This tool covers both, using an AI vision model that reads what is in the picture: subject, composition, lens, light, colours and mood.
There are three modes. Image to prompt writes a ready prompt for Midjourney, Flux, GPT Image and Veo / Kling video, for 2 credits. Product from photo writes a title, a description, five feature bullets and SEO keywords in English, Arabic or both, for 2 credits. Bulk descriptions handles up to 25 products from plain text lines, at 1 credit per 5 products. A prompt gives a close result, not an identical copy, and a description still needs a check of prices, sizes and materials. A new account gets 2 free credits once the email is confirmed.
Why use it
- A vision model that understands what is in the image, not just its colours and dimensions.
- Prompts written the way each generator likes them: Midjourney, Flux / Stable Diffusion, GPT Image / Gemini, plus video prompts for Veo / Kling.
- A complete product listing in English and Arabic in one run: title, description, five bullets and SEO keywords.
- A fixed price shown on the button before each run, and an automatic refund if a run fails.
- We do not store your images, text or results; we log only the date, mode and credits used.
How to use
Sign in
Press “Sign in and try it free” and type your email to get a one-time sign-in link, or continue with Google. New accounts get 2 credits after confirming the email.
Pick a mode
Choose one of three modes: Image to prompt with the generator you plan to use, Product from photo for a single item, and Bulk descriptions for a list of products.
Upload and check the price
Upload an image (JPG, PNG, WebP) up to 10 MB, or paste your product lines in bulk mode. The price is on the button before you press it, and in bulk mode it follows the number of lines.
Run it and review
Press the run button and wait a few seconds, then copy the prompt or the description. Try the prompt in your generator and adjust it, and read the description through before it goes on your shop.
Examples
An Instagram perfume shop in Kuwait
A photo of a perfume bottle next to dried roses. In Product from photo with the note “100 ml, 18 KWD” you get a title, a description in English and Arabic and five bullets ready for a post. The scent notes and how long it lasts are not visible in the photo, so add them to the notes instead of leaving them to the model.
An abaya store listing 20 products
Twenty lines such as “Black crepe abaya, hand embroidery, sizes 52–60”. In bulk mode with the luxury tone each abaya gets a title, a short description and bullets, for 4 credits in total. Check sizes and prices before uploading them to the store.
A designer recreating the lighting of their own photo
A designer shot their product in side window light and wants the same look on other backgrounds. In Image to prompt with the Flux target, the prompt describes the light direction, its colour, the shadows and the lens, so they reuse it and change only the background.
A video prompt for Veo or Kling
A photo of a café at sunset. With the video target the prompt describes the camera move, such as “slow dolly-in toward the window”, the steam and people moving, and the shot length, ready for Veo / Kling instead of a still-image prompt.
| Target | Best for | What the prompt focuses on |
|---|---|---|
| Midjourney | Artistic and ad images with a clear style | Dense style and mood keywords, with settings such as aspect ratio |
| Flux / Stable Diffusion | Realistic photos and fine control over detail | Descriptive sentences for subject, lens and light in a clear order |
| GPT Image / Gemini (Nano Banana) | Editing by chat and scenes that include text | Full natural description, as if briefing a designer |
| Video: Veo / Kling | Short clips for ads and reels | Camera move, subject motion, length and shot sequence |
| Product from photo | Online shops and social media pages | Title, description, five bullets and SEO keywords instead of a prompt |
Tips
- Use your own images, or images you have the rights to. Do not use the tool to copy other artists' work and people's faces without permission.
- Put what the photo cannot show in the notes: price, sizes, material and brand. The model cannot know them from the image.
- Run the prompt as it is first, then change one phrase per try so you can see what each change does.
- In bulk mode, use multiples of 5 products so every credit is fully used.
- Pick a well-lit photo that shows the whole product; a dark photo gives a less accurate description.
Your image is shrunk to 1280 px in your browser before it is sent, then travels over an encrypted connection to our server and on to the AI model provider for processing, and the result comes straight back to you. We do not keep images, text or results in our database. We record only the date, the mode, the credits used and the processing cost, so we can show your usage history and keep our accounts.
Frequently asked questions
Is the tool free?
It runs on credits, because every run costs us vision-model processing. A new account gets 2 free credits after confirming the email, enough for one run in an image mode. Trial credits are valid for 30 days.
How much do credits cost?
One credit is $0.05. Pay as you go from $5 (100 credits), or pick a pack: $10 = 220 credits, $15 = 345 credits and $25 = 600 credits. You buy them once; there is no subscription and no auto-renewal.
How many credits does one run use?
Image to prompt costs 2 credits and Product from photo costs 2 credits. Bulk descriptions cost 1 credit per 5 products, so 25 products cost 5 credits. The price is on the button before you press it.
What happens if a run fails?
Credits are taken when the run starts. If it fails for any reason on our side or at the AI provider, they go back to your account automatically and the refund appears in your history.
How is this different from your free Image to Prompt page?
The free /image-to-prompt/ page runs inside your browser, works offline and builds a prompt from image measurements such as colours, brightness and size. This tool sends the image to a vision model that understands the content: who is in it, what the product is, where the scene is and how the light falls.
Which images are accepted, and what are the limits?
JPG, PNG and WebP images up to 10 MB; your browser shrinks them to 1280 px before sending. Bulk mode takes up to 25 products per run. A daily limit protects the service from automated abuse, and it grows as you buy more credits.
Will the prompt give me the exact same image?
No. The prompt describes the image closely so you get a similar style, light and composition, but the generator draws a new image every time, so do not expect an identical copy.
How do I format products in bulk mode?
One product per line: the name, then short notes separated by commas, such as “Black crepe abaya, hand embroidery, sizes 52–60”. Then pick the language and the tone (luxury, friendly, simple), and each product gets a title, a short description and bullets.
Can it write descriptions in Arabic and English?
Yes. Choose English, Arabic, or both together in one run at the same price. Prompts are written in English because image generators follow it best, and you can type your notes in Arabic.
What are the SEO keywords for?
They are suggested search terms that fit the product, for your page title and tags, to help search engines understand what you sell. They do not guarantee any ranking, so remove any that do not fit your shop.
Do you keep the images I upload?
No. We do not keep images, text or results. We record only the date, mode, credits used and processing cost. The image passes through the AI provider to be processed.
How do I buy credits, and when do they expire?
On the My credits page, pick a pack and pay with any of these: our Gumroad store, PayPal, USDT on TRON, USDC on Solana. Purchased credits are valid for one year from purchase, and refund details are on the refund policy page.
Last updated: