By Tool

Recover the prompt from a Stable Diffusion PNG

Most Stable Diffusion interfaces write the full generation settings into a "parameters" PNG text chunk — prompt, negative prompt, sampler, steps, CFG, seed, and model.

Try it now: AI Metadata Inspector

Free, no login — upload an image and read its embedded metadata in your browser.

What to know

Popular Stable Diffusion interfaces save the generation details into a single PNG text chunk usually named "parameters". It holds the positive and negative prompt, sampler, steps, CFG scale, seed, size, and often the model name and hash — everything needed to reproduce the image.

As with ComfyUI, this lives in the PNG only. Convert the file or strip its metadata and the parameters chunk is gone. From the original PNG, recovery is straightforward.

How to recover it

  1. 1

    Use the original PNG

    The parameters chunk exists only in the PNG the interface saved — not in a converted or re-uploaded copy.

  2. 2

    Read the "parameters" chunk

    Upload the PNG to the AI Metadata Inspector to see the parameters chunk, or use your interface’s built-in PNG-info reader.

  3. 3

    Separate prompt from settings

    The chunk packs prompt, negative prompt, and settings together; the first block is the prompt, followed by "Negative prompt:" and the parameter list.

Extracting the parameters chunk on import turns a folder of PNGs into a searchable library where every prompt, seed, and model is a filter — not a chunk you have to re-read one file at a time.

Frequently asked questions

Frequently Asked Questions

Quick answers to the most common questions.

Related recovery guides

Never lose a prompt again

Numonic extracts the prompt, model, seed, and settings from every AI image on import and keeps them searchable — so recovery stops being a rescue mission and the prompt is always attached to the work.