Remove Stable Diffusion Metadata from Images
Automatic1111 stores your Stable Diffusion generation parameters (prompt, negative prompt, seed, CFG scale, sampler, model hash) in a PNG tEXt chunk with the key parameters. In JPEG and WebP exports the same data lives in the EXIF UserComment field. ComfyUI embeds the full node workflow as JSON in PNG tEXt. StripShot removes all of them at the binary level in your browser.
Quick answer
Drop your Automatic1111 image into StripShot. It detects the PNG parameters tEXt chunk (or EXIF UserComment for JPEG/WebP), strips it at the byte level, and downloads a clean file with no re-encode and no quality loss.
Stable Diffusion interfaces like Automatic1111, ComfyUI, and Forge embed detailed generation parameters into every image. This includes your full prompt, negative prompt, seed, CFG scale, steps, sampler, and model hash. StripShot removes all of it.
What the Automatic1111 parameters text chunk looks like
When you save a PNG from Automatic1111, it writes a tEXt chunk with the keyword parameters directly into the file. The value is a plain text block that anyone can read with Automatic1111's PNG Info tab, ExifTool, or any hex editor. Here is what it looks like:
a detailed fantasy portrait, masterpiece, best quality, intricate details, (bokeh:1.2), sharp focus Negative prompt: blurry, deformed, ugly, low quality, nsfw, watermark, signature Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 3782615049, Size: 512x768, Model hash: a1535d0a42, Model: v1-5-pruned-emaonly, Clip skip: 2, ADetailer model: face_yolov8n.pt
This entire block is stored in plain text inside the PNG file. StripShot removes the parameters tEXt chunk entirely at the byte level -- no re-encoding, nothing left behind.
What Stable Diffusion embeds in your images
Stable Diffusion itself is an open-source model. The metadata comes from the interfaces used to run it. Each interface stores generation data differently, but most use PNG tEXt chunks and EXIF fields.
PNG tEXt parameters chunk (Automatic1111 / AUTOMATIC1111 Web UI)
Automatic1111 is the most popular Stable Diffusion interface. It stores generation parameters in a PNG tEXt chunk with the key parameters. This chunk contains a structured text block with your complete generation recipe.
The tEXt chunk includes: full positive prompt, full negative prompt, steps count, sampler name (Euler, DPM++, etc.), CFG scale, seed value, image dimensions, model hash, model name, and any active extensions like ControlNet or ADetailer.
EXIF UserComment parameters (Automatic1111 JPEG)
When saving as JPEG, Automatic1111 stores the same generation parameters in the EXIF UserComment field. Some interfaces also write to EXIF ImageDescription and Software fields.
The EXIF data is structured identically to the PNG tEXt data. It contains the full prompt, negative prompt, and all generation settings in a parseable format. Any EXIF viewer can read it.
WebP metadata parameters (Automatic1111)
When Automatic1111 saves as WebP, it writes generation parameters into the EXIF UserComment field embedded in the WebP container. WebP files can carry both EXIF and XMP metadata blocks. Automatic1111 uses the EXIF block; some third-party scripts also write to XMP.
The WebP EXIF block follows the same format as JPEG EXIF. Your full prompt, negative prompt, seed, and model hash are stored in plain text in the UserComment field and can be read by any metadata viewer.
ComfyUI workflow data
ComfyUI embeds the entire node workflow as JSON in PNG tEXt chunks. This includes not just the prompt and settings but the complete node graph, every connected node, and all parameters for each node.
ComfyUI workflow data is typically much larger than Automatic1111 parameters. A complex workflow can add 50-200KB of metadata to the image file.
Model hash, model name, and Civitai detection
The model hash is a short identifier (like a1535d0a42) that identifies which checkpoint model was used. Automatic1111 embeds both the hash and the full model name in every image.
Anyone can look up this hash on Civitai or other model repositories to identify the exact checkpoint you used -- including custom LoRAs and fine-tuned models. This exposes your entire workflow, training choices, and model sources. Stripping removes both the hash and name.
Seed, steps, CFG scale, and sampler
These four values, combined with the prompt and model, are enough to exactly reproduce your image. The seed is a random number that determines the initial noise pattern. Steps controls iteration count. CFG scale controls how strictly the model follows the prompt. The sampler determines the denoising algorithm.
How StripShot removes Stable Diffusion metadata
- PNG files: Removes all tEXt, iTXt, and zTXt chunks. This eliminates the Automatic1111
parameterschunk, ComfyUI workflow JSON, and any custom metadata chunks. - JPEG files: Strips APP1 markers containing EXIF data with generation parameters. Removes UserComment, ImageDescription, and Software fields.
- WebP files: Removes EXIF and XMP metadata segments that may contain SD generation data.
- All formats: Binary-level stripping. No canvas re-encoding. Zero quality loss. Original resolution preserved.
Processing happens entirely in your browser. Your images never leave your device.
Does local Stable Diffusion trigger Instagram's "Made with AI" label?
Standard Automatic1111, ComfyUI, and Forge do not embed C2PA Content Credentials. Instagram's "Made with AI" label is primarily triggered by C2PA -- so by default, local SD images are less likely to be auto-labeled than images from DALL-E, Adobe Firefly, or Midjourney.
However, Instagram also uses perceptual hash matching: it compares your image against databases of known AI outputs. If your image is similar to a known AI-generated image, it may still be labeled even without C2PA credentials. This affects any image style that closely resembles common AI outputs.
Some Stable Diffusion extensions (including third-party C2PA signing scripts) can add C2PA credentials to SD images. If you use any such extension, your images will trigger the AI label just like cloud-based tools.
StripShot's anti-fingerprint mode (Pro) applies invisible pixel-level changes that completely alter the perceptual hash of your image, defeating hash-based detection without any visible quality change.
Why removing SD metadata matters
Unlike DALL-E and Midjourney, Stable Diffusion does not use C2PA Content Credentials. Social platforms are less likely to auto-label SD images. But the generation parameters are a bigger problem for a different reason: prompt theft.
Anyone who downloads your image can open it in Automatic1111 (using the PNG Info tab) and see your exact prompt, negative prompt, model, seed, and settings. They can reproduce your image pixel-for-pixel. For artists selling AI art, this is a direct business risk.
SD metadata also reveals which model you used. If you trained a custom LoRA or used a specific Civitai checkpoint, the model hash and name are visible in the metadata. Stripping this protects your model choices and training investments.
Forge and other SD interfaces
Stable Diffusion Forge (a popular Automatic1111 fork) uses the same PNG tEXt metadata format. InvokeAI stores generation data in its own metadata format but also writes to PNG tEXt chunks. SD.Next and vladmandic forks follow similar patterns.
StripShot removes all PNG tEXt chunks regardless of the interface that created them. You do not need to know which SD interface was used. All generation metadata is stripped.
Frequently asked questions
Can someone steal my Stable Diffusion prompt from my image?
Yes. Anyone who downloads your SD-generated image can open it in Automatic1111's PNG Info tab and see your exact prompt, negative prompt, seed, and all generation settings. The metadata is stored in plain text. Stripping it with StripShot removes all generation parameters permanently.
Can you recover the seed from a Stable Diffusion image after removing metadata?
No. Once the PNG tEXt chunk or EXIF UserComment field is stripped, the seed value is gone permanently. There is no way to recover a seed from a stripped image. The pixel data is unaffected, but the generation parameters -- including seed, prompt, model hash, and settings -- are deleted at the binary level and cannot be reconstructed.
Can Civitai identify which model I used from the image metadata?
Yes. Automatic1111 embeds the model hash (e.g., Model hash: a1535d0a42) and model name in the metadata of every saved image. Anyone can look up this hash on Civitai or other model repositories to identify your exact checkpoint. Stripping removes both the hash and the model name.
Does Stable Diffusion embed C2PA Content Credentials?
No. Unlike Adobe Firefly and DALL-E, Stable Diffusion does not use C2PA Content Credentials by default. SD metadata is stored in PNG tEXt chunks and EXIF fields, not JUMBF containers. Standard local SD images are less likely to trigger Instagram's "Made with AI" label, but the generation parameters are still exposed in plain text.
Does removing SD metadata affect image quality?
No. StripShot removes metadata at the binary level without re-encoding the image. The pixel data is untouched. Your image retains full resolution and original quality after stripping.
How do I remove ComfyUI workflow data from an image?
ComfyUI embeds the entire node workflow as JSON in PNG tEXt chunks, which can add 50-200KB of metadata. StripShot removes all tEXt, iTXt, and zTXt PNG chunks, eliminating ComfyUI workflow data completely. Drop the image into StripShot and it handles the rest.
Related guides
Remove Midjourney Metadata
Strip Midjourney generation parameters and prompt data.
Remove Flux and Fooocus Metadata
Strip Flux AI generation parameters and PNG info chunks.
Remove DALL-E Metadata
Strip OpenAI EXIF tags and C2PA Content Credentials.
Remove All AI Metadata
Complete guide covering every major AI image generator.
Bypass Instagram AI Detection
Remove the "Made with AI" label on Instagram posts and Reels.
Bypass AI Detection
Platform-by-platform guide for Instagram, Pinterest, Facebook.
Pricing
Clean 3 images free. Then go unlimited.
No card needed to start. Upgrade anytime.
- 2 images/day free
- 1 video/day free
- AI detection (70+ signatures)
- 100% private, no uploads
- Unlimited image cleanings
- 10 videos/day included
- Batch processing
- Smart Strip & hex viewer
- Edit & write EXIF tags
- Unlimited image cleanings
- Unlimited video cleanings
- Strip Sora, Runway, Kling
- MP4 + MOV support
- New generators added regularly