PNG Info

Reads the text chunks stored inside a PNG. For images made with Stable Diffusion or ComfyUI that means the full prompt, the negative prompt, the seed, the sampler and the model name.

This tool does not upload your files. Close the tab and nothing is left behind.

Drag & drop files here, or

Choose files

Supported: PNG

Recovering the prompt from a generated image

Anyone who generates images in bulk knows the feeling of opening a folder from three weeks ago and having no idea which prompt produced the one good result. Stable Diffusion WebUI, ComfyUI and InvokeAI all quietly write the generation settings into the PNG itself, in a text chunk that travels with the file. This tool reads those chunks and splits them into the prompt, the negative prompt, the seed, the sampler, the CFG scale and the model name.

Each field has its own copy button, so you can lift just the prompt and feed it straight back into your pipeline, and the PNG header values such as dimensions and bit depth are listed separately underneath. The catch is how easily that record is destroyed. Converting the file to JPG removes it, re-saving it in an image editor removes it, and downloading it back from a site that strips metadata removes it too, which is why images found online so often come back empty.

Some services never write the prompt in the first place: images from Midjourney or DALL-E carry nothing to read. The reverse is worth knowing as well, because sharing your own PNGs means shipping your prompts to whoever downloads them. Running them through the Remove EXIF tool deletes the text chunks with no loss of quality. Either way the file is read inside your browser and is never sent anywhere.

How to use

  1. 1

    Drop in a PNG file.

  2. 2

    Any generation data is split into labelled fields.

  3. 3

    Use the copy button next to a field to take just the prompt.

Frequently asked questions

Which images carry this?
Original PNGs from Stable Diffusion WebUI (A1111), ComfyUI, InvokeAI and similar tools. They write the generation settings straight into the PNG's tEXt chunk.
Why do I get nothing?
The chunk is lost when the file is converted to JPG, re-saved by an editor, or downloaded back from a site that strips it. Some services, such as Midjourney and DALL-E, never write the prompt into the file at all.
How do I stop my own PNGs from carrying the prompt?
Run them through the Remove EXIF tool, which deletes the text chunks without any loss of quality, before you share them.

Related tools