快速入口: 本页是 AI Image Inspector 的专注入口 Midjourney. 前往 AI Image Inspector→
Drop a PNG, JPEG, or WebP image
AI generation fingerprints are detected in your browser. Nothing is uploaded.
Check if an Image is from Midjourney — Free Midjourney Image Detector
预期用途
- The inspector reports fingerprints it finds in the file. It does not prove the image is real when no fingerprints are present — a re-encoded screenshot of a Midjourney image will also return 'No AI generation traces found.'
- Screenshots of Midjourney images typically strip the PNG text chunks and the EXIF block. If the original file is no longer available, no structural detector can recover the fingerprint.
使用方法
Drop the image you want to test
Drag a PNG, JPEG, or WebP file from your desktop, your browser, a Midjourney gallery, or a downloaded folder into the drop zone. The image is read into a Uint8Array in your browser. Nothing is uploaded — the detection runs as pure JavaScript in the same tab.
Read the Midjourney fingerprint report
The inspector checks for Midjourney-specific signals: PNG tEXt chunks that Midjourney's Discord bot writes during the upscale step ('Midjourney' in Software or Description), EXIF Software and Creator tags, and profile cues common to Midjourney generations. Each detection is shown as a card with a 10-bar probability meter and the byte-level evidence. The highest-confidence card is the answer.
Verify and clean up
If the inspector reports a Midjourney fingerprint, open the Strip Image Metadata tool to wipe the PNG text chunks and the EXIF block. Re-run the cleaned file through the inspector to confirm the score has dropped below the detection threshold.
常见问题
How can I check if an image is from Midjourney?
The fastest way to check if an image is from Midjourney is to drop it into the AI Image Inspector on this page. The inspector reads the PNG text chunks and EXIF Software tag that Midjourney's Discord bot writes during the upscale step, and reports each detection with byte-level evidence. The whole check runs in your browser — no upload, no server, no account.
Does Midjourney put a watermark on its images?
Midjourney does not embed a C2PA manifest or an invisible pixel-level watermark by default. Instead, the Discord bot writes identifying information into PNG text chunks during the upscale step (most commonly in a Software or Description chunk that names 'Midjourney'). These chunks survive downloads from the Midjourney web gallery and Discord. Older Midjourney v4 outputs sometimes also carry an EXIF Software tag.
Will this work for Midjourney v5, v6, and v6.1?
Yes. Midjourney v5, v6, and v6.1 all write the same style of PNG text chunks during the upscale flow. The Discord bot adds them consistently across versions, so the inspector reports them with the same confidence. The key trade-off is that Midjourney v5+ rarely adds anything beyond the basic text chunk — the detector relies heavily on the chunk being preserved through any download and re-encode.
What if the image was screenshotted or downloaded from a third-party gallery?
A screenshot or a re-encoded JPEG from a third-party site typically strips the PNG text chunks and the EXIF block, so the inspector will report 'No AI generation traces found.' Screenshots and re-encodes are a known limitation of every structural detector — the only way to keep the Midjourney fingerprint is to keep the original PNG file that the Discord bot produced.
Can this tell Midjourney apart from DALL-E 3 or Stable Diffusion?
Yes. The three vendors leave three different fingerprints. DALL-E 3 (ChatGPT) leaves a C2PA Content Credentials manifest naming OpenAI. Stable Diffusion leaves a 'parameters' or 'workflow' text chunk with the prompt and sampler config. Midjourney leaves a 'Midjourney' text chunk and minimal other metadata. The inspector reports each vendor independently and the highest-confidence card names the actual source.
Is this a Midjourney image detector?
It is a structural Midjourney image detector. It reads the metadata, the PNG text chunks, and the EXIF software tag that Midjourney's Discord bot leaves on its files. It does not run a neural network on the pixels, so it will not catch a Midjourney image whose metadata has been deliberately scrubbed. For any image that still carries the PNG text chunks, the answer is unambiguous.
What does the verdict mean?
The verdict is the aggregate answer across all detectors in the inspector. 'No AI generation traces found' means none of the detectors produced a non-zero score — the image does not carry any known Midjourney, DALL-E, Stable Diffusion, or other AI fingerprint. 'Likely generated by a single AI vendor' means one detector produced a high score — the card at the top names the vendor. 'Some weak signals detected' means at least one detector produced a low score — inconclusive, often a false positive on heavily-edited images.
Are my images uploaded to a server?
No. The image bytes are read into a Uint8Array in your browser. The PNG chunk reader, the EXIF reader, and every other detector all run as pure functions in JavaScript. No network call, no server roundtrip, no third-party API, no telemetry, no account. The file never leaves your device, and nothing is stored between sessions. You can verify this by watching the Network tab in DevTools while you drop a file in.
限制说明
- Structural detection, not pixel-levelThe inspector reads metadata, PNG text chunks, EXIF tags, and XMP fields. It does not run a neural classifier on the pixel content. A Midjourney image that has been re-encoded with all metadata stripped will be reported as 'No AI generation traces found.' The inspector catches any Midjourney image that still carries at least one of the fingerprints — it cannot catch a fully-cleaned Midjourney image that has been re-encoded.
- Screenshots defeat detectionWhen you take a screenshot of a Midjourney image, your operating system re-encodes the pixels and typically drops the PNG text chunks and the EXIF block. The screenshot will be reported as 'No AI generation traces found.' The only way to keep the fingerprint is to keep the original PNG file that the Discord bot produced (or the file you downloaded from the gallery that preserved it).
- PNG, JPEG, and WebP onlyThe PNG chunk reader works on PNG tEXt/zTXt/iTXt chunks. EXIF and XMP are read from JPEG APP1, PNG tEXt, and WebP 'EXIF'/'XMP ' chunks. AVIF, HEIC, and TIFF use different containers and are not decoded by this inspector. If your file is in one of those formats, convert it to PNG or JPEG first.
平台说明
- macOS
- Preview.app and Quick Look do not surface the PNG text chunks that Midjourney writes. The inspector is the right tool for diagnosing the provenance of a Midjourney image dropped from Safari, the Midjourney web gallery, or a Discord chat. To inspect a specific tag, right-click the file → Get Info → More Info shows EXIF but not PNG text chunks.
- Windows
- Windows Photos and File Explorer show EXIF and basic XMP, but not PNG text chunks. Right-click → Properties → Details shows the Software tag if it is set, but the Midjourney text chunks are not visible. For full Midjourney provenance, drop the file into the inspector — the report covers every Midjourney fingerprint in one pass.
- Linux
- CLI alternative for some Midjourney fingerprints: `exiftool -api RequestAll=3 image.png` shows the PNG text chunks and EXIF tags, and will reveal the 'Midjourney' description. It does not run profile or pixel-cue detectors, but those are the only signals available structurally. The inspector is the right pick when libexiftool is missing PNG chunk support or when you want a single report that covers every fingerprint.
- Web
- Runs entirely client-side. The image is read as a Uint8Array, the PNG chunk reader, EXIF reader, and XMP reader all run as pure functions, and the report is rendered as HTML. No upload, no server roundtrip, no telemetry, no third-party calls. The file never leaves your device.