Free
$0
Free credits every day, more with a free account.
- All four prompt formats
- Image breakdown
- Prompt history with an account
- No card needed
Free image to prompt generator
Image to prompt,
written for your model
Upload a picture and get a ready to paste prompt for Midjourney, Nano Banana, Flux, Stable Diffusion or ChatGPT, with the lighting, lens and palette spelled out.
Prompts that already speak your model's language
Good prompts are specific. genprompt names the things you would miss: the focal length, the light setup, the exact palette.
Four prompts from one upload
Universal, Midjourney, Nano Banana and Flux are written in a single pass. Switching format is instant and free.
Written in each model's syntax
Midjourney gets its flags and a no list. Flux gets a full negative prompt and guidance. Gemini gets a precise paragraph.
A breakdown you can edit
Subject, lighting, lens, palette and mood are listed apart, so you can change one thing and keep the rest.
Your image stays yours
It is resized in your browser, read once by the model and never saved on our servers.
Stop guessing keywords. Start from the image you already love.
Three steps, under thirty seconds
- 01
Drop an image
A photo, a render, an ad you like. JPEG, PNG or WebP.
- 02
We read it
Style, medium, lighting, lens, palette and any visible text are pulled out.
- 03
Paste the prompt
Pick your model, copy, and generate in the tool you already use.
An image to prompt generator for every model
Each model reads prompts differently. Open the guide for the one you use.
Image to Midjourney prompt
Upload any image and get a Midjourney prompt with the right flags: aspect ratio, style raw, stylize and a no list. Free to try, no card needed.
Image to Nano Banana prompt
Turn any image into a detailed Nano Banana prompt for Gemini image models, including Nano Banana Pro. Scene, subject, light and camera in one precise paragraph.
Image to Flux prompt
Upload an image and get a detailed Flux prompt with style and quality tokens, a negative prompt and suggested guidance and steps. Works with Flux in ComfyUI and online.
Image to Stable Diffusion prompt
Get a Stable Diffusion prompt from any image, with a negative prompt, CFG scale and steps. Ready for SDXL, Automatic1111, Forge and ComfyUI.
Image to prompt for ChatGPT
Turn any image into a clear prompt for ChatGPT and GPT Image. Get a natural description of subject, style, light and camera you can paste into a chat.
Text to prompt
No image? Write a rough idea or paste a prompt from another tool and get it for every model.
One prompt is a starting point
Every result keeps its breakdown, so you can build on it without uploading again.
Prompt X-ray
See which words describe which part of the image. Click a zone to change only that part.
Same style, new subject
Keep the look, lighting and lens. Swap what the image shows.
Remix one attribute
Change only the lighting, mood, lens or palette and keep the rest.
Variants
Faithful, creative and minimal takes on the same image.
From text
Turn a rough idea, or a prompt from another tool, into prompts for every model.
Style DNA
Extract the style shared by 3 to 5 references and reuse it on any subject.
Render and compare
Render the prompt, get a match score against your original, apply the fixes.
Batch
Up to 20 images in one go, exported as CSV or JSON.
Start free, pay when it earns its place
No card needed to try it. One image is one credit, every format included. See what each credit buys
Credit packs
From $9
Pay once, credits never expire. Needs a free account. Prices exclude VAT, added at checkout where applicable.
300 credits
$9 · 3¢ per prompt
800 credits
$19 · 2.38¢ per prompt
2,000 credits
$39 · 1.95¢ per prompt
Questions
How do I get a prompt from an image?
Drop the image in the box at the top of this page. In a few seconds you get a breakdown of the image and a prompt for each model. Pick your model tab and copy.
Is the image to prompt generator free?
Yes, you get free credits every day with no account and no card, and more when you create a free account. One image is one credit and includes every model format.
What does genprompt do?
You upload an image and get a prompt that describes it the way image models understand: subject, composition, style, lighting, lens and palette. Paste it into your generator to get something close in look and feel.
Which image models are supported?
Every upload gives you a universal prompt plus versions tuned for Midjourney, Nano Banana and Gemini, and Flux or Stable Diffusion. The universal one works with most other models.
Do you store my images?
No. The image is resized in your browser, sent once to the vision model for analysis and not saved on our servers. When you are signed in, only the generated text is kept in your history.
Will the prompt recreate my image exactly?
No prompt can promise a pixel match. You get the same subject, framing, style and mood, and image models add their own variation on top.
Can I upload images of people?
Yes, but people are described rather than named. Only upload images you have the right to use.
What files can I upload?
JPEG, PNG and WebP up to 20 MB. Large images are scaled down to 1536 pixels before they leave your device.
How do credits work?
One uploaded image uses one credit and gives you all formats at once. Switching between Midjourney, Flux and the others does not use more credits.
Your next prompt is
one upload away
Free to try, no card, and your image is never stored.
Upload an image