Remove Stable Diffusion & ComfyUI PNG Metadata
Slipp bilder her eller klikk for å laste opp
JPG, PNG, WebP, AVIF, HEIC · Max 15 MB · Batch 30 · Geen GIF
ComfyUI graphs and A1111 seeds live inside PNG text chunks — not just EXIF.
Recommended workflow: metadata checker → remover
Each AI exporter leaves different fingerprints in your file. Inspect read-only before you strip fields — then use the remover on this page for the upload you will publish.
- Step 1 — inspect your file: AI metadata checker.
- Upload your export to the AI metadata checker (read-only, browser-only)
- Review C2PA, XMP, PNG chunks, and EXIF software strings
- Use the remover above with matching options enabled
- Download the cleaned JPG, PNG, or WebP
- Upload to Instagram, Facebook, Pinterest, TikTok, or LinkedIn
After Photoshop, Canva, or compositing
A raw AI download is not automatically upload-ready. Generative Fill, Canva AI backgrounds, or compositing can add C2PA or IPTC AI fields on top of the original export. Always clean the final flat raster you publish — not intermediate layers or PSD masters.
Why SD / ComfyUI PNGs trigger AI labels
Unlike a camera RAW pipeline, diffusion exports often carry full workflow serialization inside the PNG container: node graphs, prompts, seeds, and checkpoint names.
That is valuable for reproducibility — and it is also a loud automated signal for "this file came from a generator". Inspectors and community tooling have long surfaced PNG chunk data on creative uploads.
A1111 (SD Web UI) vs ComfyUI vs headless inference
- SD Web UI (A1111) — classic parameter + PNG chunk patterns; watch embedded PNG info when "Save metadata"-style options are on.
- ComfyUI — workflow JSON in PNG text chunks is the default mental model.
- API / batch workers — confirm your writer does not reintroduce chunks after post-processing.
How this differs from Midjourney and DALL·E
Each toolchain emphasizes different metadata classes:
- Midjourney leans on XMP.
- DALL·E leans on C2PA.
Stable Diffusion front-ends lean on PNG ancillary text. The remover above hits all three classes so you do not need separate desktop apps per toolchain.
Recommended cleaning checklist
- Flatten to the final raster (PNG or JPG) you will ship.
- Enable PNG text chunk removal for PNGs from SD stacks.
- Keep XMP enabled if you passed through another editor afterward.
- Keep C2PA enabled if you mixed in Firefly or ChatGPT image assets.
- Confirm with the metadata checker.
Deep dive on the blog
- Read stripping Stable Diffusion metadata for chunk-level detail and batch ideas.
ComfyUI shortcut
Looking for ComfyUI-only wording? We keep a /comfyui URL that redirects here so bookmarks and nav stay stable.
What this tool removes (SD / ComfyUI)
- PNG tEXt / iTXt workflow and parameters chunks
- XMP generation blocks when present
- C2PA when mixed from other tools
- EXIF software strings
Does not remove pixel watermarks. See disclaimer .
Clean this file before you upload
Strip C2PA and XMP in your browser, then publish the cleaned copy.
Local SD pipelines and social upload
Self-hosted SD users assume «local» means «no labels» — platforms only see the uploaded bytes. ComfyUI graphs belong in archives, not in PNG chunks attached to Pins.
Civitai model cards and preview images
Model authors reuse sample images — upload-prep before using Civitai previews as your own Instagram posts without checking licenses and metadata.
SD for architectural viz
Archviz artists post AI-enhanced renders on LinkedIn — clients compare before/after; metadata strip avoids «AI» on photoreal renders.
SD for sticker shops
Sticker PNGs with transparency need chunk removal — alpha does not hide workflow JSON.
Hugging Face Spaces demos — download from Space, strip before using output in commercial Twitter ads.
SD lab notebook hygiene
Lab notebook records strip timestamp for reproducibility separate from PNG params.
Conference demo strip all grid cells before audience phones post.
Local SD stack → checker → remover → marketplace upload
Sellers and meme makers search «remove ComfyUI metadata before Instagram» when automatic labels appear on SD outputs that look like ordinary photos. PNG text chunks and XMP sit beside the bitmap — platforms read them at ingest.
Workflow: export final PNG from ComfyUI or A1111 → optional checker → strip chunks/XMP/C2PA → rename neutrally → upload to Instagram, Etsy, or Reddit OC posts.
- Save Image from ComfyUI or PNG from A1111 — not a Discord re-share
- Checker: note PNG text chunks vs XMP parameter blocks
- Remover with PNG chunk, XMP, and C2PA options enabled
- Download to «sd-upload-ready» separate from workflow JSON archives
- Upload cleaned raster to social or marketplace attachment
Batch up to 30 SD PNGs per browser session.
Checkpoint archives vs publish masters
Keep workflow JSON and credentialed PNGs in a «comfy-archive» tree for reproducibility. Publish only flattened, cleaned masters to Pinterest and Instagram. Document internally which checkpoint produced each listing hero when clients audit campaigns.
Myth: PNG chunks are «developer only.» Reality: Instagram, Pinterest, and Etsy ingestion pipelines parse them on many uploads — strip before attach.
SD meme pages and OC photography subs
High-karma Reddit subs run automod on SD workflow chunks — same strip discipline as Instagram. Keep «sd-meme-ready» folder separate from «comfy-workflows» JSON tree.
Etsy digital download shops selling SD art packs should strip preview PNGs before Pinterest promos — buyers see clean files; promos must match.
Civitai release day strip checklist
Model release posts need strip on every sample PNG in announcement thread — ML Twitter QT chains spread uncleared workflow chunks by afternoon.
Keep «sd-release-ready» separate from «comfy-workflows» JSON tree on release day.
FAQ
Where does ComfyUI hide metadata in PNG files?
ComfyUI and many SD Web UI (A1111) builds store serialized workflow graphs inside PNG text chunks (for example parameters or workflow JSON). Those chunks can survive re-encode steps in some tools, so explicitly strip PNG text metadata before social upload.
Does JPG export avoid the problem?
JPG cannot store the same PNG tEXt workflow payload, but exporters may still attach XMP or EXIF software fields. Always clean the exact file you upload.
Is removing PNG chunks lossless for pixels?
Yes. Removing ancillary text chunks does not alter decoded pixel values.
I saved as JPG from Comfy — am I done?
JPG avoids PNG workflow chunks but may still carry XMP/EXIF. Run the final JPG through the same pipeline.
Does latent upscaling remove chunks?
Not reliably. Some nodes re-encode PNG and drop chunks; others preserve them. Never assume — verify with the checker.
ComfyUI workflow JSON inside PNG chunks
ComfyUI and A1111 exports frequently embed workflow JSON in PNG text chunks — Instagram and Pinterest parsers may read these blocks. Enable PNG chunk removal alongside C2PA. AUTOMATIC1111 infotext in PNG is a separate signal from C2PA — strip both classes of hidden data.
SDXL and LoRA merge exports
Custom checkpoints do not remove metadata writers — your PNG still needs upload-prep. Batch generate 30 finals per session, clean, then schedule social posts.
SD Forge and alternate UIs
Forge and SD.Next write different PNG parameter formats — enable both chunk and C2PA removal to cover forks without retuning per UI.
ControlNet and depth map chains
Multi-pass ControlNet merges stack metadata — final save node output is the only file to clean, not intermediate depth previews.
SDXL Turbo and fast samplers
Fast samplers still write PNG chunks — speed does not skip upload-prep.
SD cluster outputs for agency walls
Agencies running local SD on render farms should sync strip SOP to farm operators — uncleaned PNGs on shared NAS propagate to account managers. Checkpoint Friday updates: re-checker one image per checkpoint before Monday client sends.
SD inpaint outpaint chains
Outpaint chains add metadata each expansion — strip only final canvas export, not intermediate outpaint steps saved for debugging.
SD model release marketing
Checkpoint creators marketing new SD models post sample grids — strip each sample before Twitter grid upload; grids multiply metadata issues sixfold. Civitai preview downloads used in announcement need strip before mirror posts.
SD meetup group demo nights
Meetup demo nights post grid to Reddit — strip all demo PNGs before live audience posts.
SD strip before model card?
Yes — Civitai sample strip before model card preview upload; preview is first impression for downloads.
ComfyUI Save Image vs screenshot — which carries workflow JSON?
Save Image nodes embed PNG text chunks with full node graphs, seeds, and checkpoint names — screenshots drop metadata but destroy ad quality. Always strip the original ComfyUI PNG before Instagram or Etsy upload; enable PNG chunk removal alongside C2PA and XMP when merging with Firefly plates.
Batch-cleaning SD outputs for a Shopify hero set
Process up to 30 ComfyUI or A1111 PNGs per browser session into «shop-ready» and «archive» folders. One uncleared workflow chunk on the hero SKU can trigger GenAI flags across Pinterest catalog Pins tied to that product.
Automatic1111 metadata vs ComfyUI chunks
A1111 often writes generation parameters into PNG parameters text or XMP; ComfyUI prefers tEXt/zTXt chunks. Both are invisible beside the bitmap. The checker lists which channel your export uses so you enable the right remover toggles.
Why Etsy still flags after SD cleanup
If checker reads clean on upload bytes, triggers may be listing copy, mockup re-exports from Printify, or platform classifiers — not PNG chunks in your master. Re-clean the JPEG Printify hands to Etsy, not an older ComfyUI PNG from last month.
LoRA merge pipelines and double metadata layers
Stacking LoRA outputs in Photoshop or Firefly after ComfyUI generation can add Adobe C2PA on top of SD chunks. Flatten once, strip once on the final JPG attached to social — not on intermediate PNGs you keep for re-edit.
WebUI API responses and headless export hygiene
Headless ComfyUI queues feeding CMS uploads must checker every binary response — API wrappers sometimes strip chunks, sometimes preserve full workflow JSON. Treat each automated export as untrusted until verified.
SD strip before Civitai model preview repost?
Model authors reposting ComfyUI sample grids to Instagram should strip each grid cell export — workflow PNG chunks trigger GenAI flags on portfolio accounts even when the post is «just a preview».
RunPod / Vast headless queue strip SOP?
Automated ComfyUI queues feeding Discord bots must strip before bot posts to Reddit — bot code rarely removes PNG chunks; add strip step in post-processing script or manual QA on 10% sample.
SD strip before Hugging Face model card sample grid?
Yes — model card sample PNGs need strip before tweeting model release thread; workflow chunks trigger synthetic labels on ML Twitter.
RunPod headless queue strip SOP?
Automated ComfyUI queues feeding Discord bots must strip before bot posts — bot code rarely removes PNG chunks; QA 10% sample weekly.
SD strip before Etsy digital download preview?
Yes — Etsy preview PNGs need strip before listing publish; buyers screenshot previews to Pinterest within days and inherit workflow chunks from uncleared ComfyUI exports. Keep «etsy-sd-ready» separate from workflow archive on listing day. Spot-check first preview in checker before bulk strip runs today.
Try it now
Confirm PNG chunks are gone with our AI metadata checker.