Read AI image metadata
Find prompts, settings, and workflow data saved in a PNG image.Free to use in your browser, with no sign-in.
Drop an image hereor click to choose a PNG file. Up to 20 MB.
How to read AI image metadata
1. Choose a PNG file
Choose an AI-generated PNG up to 20 MB, or drop it onto the reader.
2. Read the metadata
Each text entry appears with its label, such as the prompt, settings, or workflow.
3. Copy the result
Click Copy result to copy all entries with their labels. Use the prompt or workflow text you need.
FAQ
Text stored inside a PNG file, such as prompts, generation settings, and workflows. Each entry appears with its label, and JSON values are formatted for reading.
Stable Diffusion WebUI and ComfyUI can save prompts and settings in PNG files. ComfyUI can also include workflow data. What you find depends on how the image was saved.
Labels are set by the app that saved the image. Stable Diffusion WebUI usually saves the prompt and settings as parameters. ComfyUI saves its prompt graph as prompt and the editor workflow as workflow. Other text stored in the PNG also appears with its own label.
The image may have been saved without metadata, or an app may have removed it during editing or sharing. This tool reads stored text. It cannot recover a missing prompt from the image.
Many websites and chat apps re-encode or convert images when you upload them, which can remove text stored in the file. To keep the prompt, share or save the original PNG file.
No. This tool reads text stored in PNG files only.
If the file includes a ComfyUI workflow, copy that JSON value into a .json file and open it in ComfyUI. You may need the models and custom nodes used in the workflow.
Browse AI image references with their prompts on ctrlc.ai.
No. Your image is processed on your device and is not sent to our servers.
No. All image tools are free to use, with no sign-in.
