A photograph that looks fine on screen can still be unnecessarily large for an application form or a product listing. This compressor creates a new image on your device. You can control the visual quality yourself, or let it search for an output that fits your chosen byte budget.
How to use this tool
- Choose one or more JPG, PNG, or WebP files.
- Select quality mode or enter a target in KB.
- Run the job, inspect the preview, and download the result.
Choose a sensible target
Start with the receiving website’s requirement rather than an arbitrary reduction. A target of 100 KB means 100,000 bytes here. Target mode searches for the highest available quality below the limit. If quality adjustments are insufficient, it reduces the pixel dimensions. The download shows the resulting dimensions and actual file size.
Keep detail where it matters
Open the preview and inspect faces, small lettering, and sharp edges. A photograph usually tolerates stronger compression than a screenshot of a spreadsheet. Increasing the target often preserves more useful detail than repeatedly saving the same heavily compressed image. Work from the original whenever you have it.
Transparency and output choices
JPG has no transparent channel, so transparent areas become the selected background color. WebP can preserve transparency. PNG is available in quality mode, but its encoder does not respond to a photographic quality slider. A smaller PNG may require changing dimensions rather than lowering a percentage.
Understand the safeguards
Select up to twenty supported images per job. The combined image budget is 40 MB, with a 24-million-pixel decoding safeguard per image. These limits protect browser memory; they are not a daily allowance. Batch outputs can be downloaded together, including files that originally had the same name.
Common questions
Will the original be overwritten?
No. Processing creates a separate download. Your source files remain in their original folder, and you decide whether to keep the new copies.
Can every image fit an extremely small target?
No useful quality can be guaranteed at every size. If the target cannot be reached within the supported reductions, the tool asks you to increase it.
Does this remove location metadata?
Re-encoding omits EXIF metadata from the new image. It does not hide addresses, faces, or other information visible within the picture.