What denoising strength should I use to remove a watermark?

For solid logos and text on photos, 0.6-0.7 denoising regenerates the masked area enough to replace the watermark. For illustrations, 0.5-0.6 is usually enough. Lower values leave traces; higher values may change the underlying content too much.

Why use 'Inpaint only masked' instead of whole image?

'Only masked' crops to the masked region and inpaints it at full resolution, giving sharper reconstruction of small areas like a corner watermark. Whole-image inpainting downsamples the region and produces softer results.

How do I fix JPEG compression artifacts?

Treat them differently from watermarks. Use a low denoising strength (around 0.3) so you smooth blockiness without regenerating content, and pair it with an upscaler at 0.2-0.3 denoise. High denoise on compression artifacts changes the image rather than cleaning it.

Is it legal to remove watermarks?

Removing a watermark from an image you do not own or have a license to use can infringe copyright and may violate the source's terms. This guide is for cleaning your own AI generations and licensed assets. Always confirm you have the right to edit and use the image.

AI Watermark & Artifact Removal Guide

Removing watermarks and artifacts with inpainting

AI-generated and processed images frequently pick up unwanted marks: a residual watermark from a reference image, stray text the model hallucinated, or JPEG compression blocks from a low-quality source. Stable Diffusion’s inpainting mode is the most reliable way to clean these up, but the right settings depend heavily on what you are removing and on which model you run. This tool gives you a tailored configuration for each case.

How it works

You select the artifact type, the image type, and your available model. The tool then recommends a denoising strength, mask blur, only-masked padding, fill mode, and step count. Text and logo removal needs high denoising (0.55-0.7) so the region is genuinely regenerated; compression-artifact cleanup needs low denoising (around 0.3) so structure is preserved while blockiness is smoothed. Padding and mask blur are tuned so the reconstructed region blends seamlessly into its surroundings.

Mask strategy and tips

Mask generously. Include a small margin around the artifact so the model has surrounding pixels to reconstruct a plausible fill.
Use only-masked at full padding. This inpaints the cropped region at high resolution, which matters most for small corner watermarks.
Iterate. A faint ghost after the first pass usually clears with a second pass on a fresh, slightly tighter mask.
Different rules for compression. Do not crank denoising for JPEG blocks — low denoise plus an upscaler restores detail without changing the picture.
Stay within your rights. Only edit images you own or are licensed to use.