Remove the Gemini Watermark From AI Images
Google Gemini stamps generated and edited images — including the popular 'nano banana' image model — with a small visible mark, typically the Gemini sparkle in a corner. It's a static, single-position overlay, which makes it one of the simplest AI watermarks to clear: one frame, one corner, no motion to track. WatermarkRemoverAI inpaints the sparkle out and rebuilds the corner so your generated image looks unbranded.
How to remove the Gemini watermark
Upload the Gemini image
Add the PNG or JPG you generated. Locate the sparkle/Gemini glyph — usually tucked in a bottom corner.
Brush the sparkle mark
Paint a tight mask over just the glyph. Since it's small and static, keep the brush close so surrounding detail is kept.
Download clean
Export the image with the corner reconstructed — no visible Gemini mark.
Good to know
This removes only the visible Gemini glyph. Google also embeds an invisible SynthID provenance watermark, which is a separate signal this tool does not target.
Frequently Asked Questions
Where is the Gemini watermark on an image?
It's a small visible glyph — the Gemini sparkle — placed in a corner of the generated image. It doesn't move or tile, so a tight mask on that one spot removes it and the rest of the image is untouched.
Does this cover 'nano banana' image edits too?
Yes. Images from Gemini's image model (nicknamed 'nano banana') carry the same style of corner mark. The removal is identical: brush the glyph, rebuild the corner.
What about the invisible SynthID watermark?
Google also embeds an invisible provenance watermark (SynthID) in the pixels. This tool removes the visible logo only; it doesn't target or guarantee removal of invisible provenance signals, which are a separate technology.
Will removing the mark hurt the rest of the image?
No. Only the masked corner is rebuilt. Because the mark is small and the surrounding pixels are intact, the reconstruction is essentially invisible.
Is it okay to remove the Gemini watermark?
Do it with images you generated for your own use. Bear in mind many platforms request that AI-generated images stay disclosed; removing a corner logo doesn't change the image's AI origin.
Related Guides
Automate it with the API
Send a file, get a job id back, poll for the clean result. Same AI engine as the web tool, in two requests.
curl -X POST https://api.watermarkremoverai.com/v1/watermark/remove-image/ \
-H "Authorization: YOUR_API_TOKEN" \
-F "files=@photo.jpg" \
-F "auto_detect=true"
# => {"uuid": "..."}
# Poll for the result
curl -X POST https://api.watermarkremoverai.com/v1/watermark/results/ \
-d "uuid=..."
import requests
API = "https://api.watermarkremoverai.com/v1/watermark"
headers = {"Authorization": "YOUR_API_TOKEN"}
with open("photo.jpg", "rb") as f:
job = requests.post(API + "/remove-image/",
headers=headers,
files={"files": f},
data={"auto_detect": "true"}).json()
result = requests.post(API + "/results/",
data={"uuid": job["uuid"]}).json()
print(result)
Prefer a manual mask? Send mask=@mask.png instead of auto_detect, or pass mask_x / mask_y / mask_width / mask_height (add mask_is_percentage=true for percentages). Optional dilation=5 grows the masked area.
Ready to clean it up?
Upload, mark the area, and download the clean result. No account needed to try it free.
Open the Tool