How to compress an image
- Add your image Drop a file, choose one from your device, or paste from the clipboard.
- Set a target size Use the preset or enter your own limit in kilobytes.
- Download the result The tool returns the best quality that fits under the limit.
Why JPG files get so large
A modern phone camera can produce a 12–48 megapixel JPG. Even with efficient compression, that is several megabytes per photo. The extra pixels are useful for printing, but most upload forms and websites do not need them.
SizeUnder decodes the JPG, corrects its orientation, and re-encodes it at a lower quality and, if needed, a smaller size. The result is still a standard JPG that opens in any photo app.
Quality, orientation, and metadata
JPG compression is lossy: once detail is removed, it cannot be restored. The tool searches for the highest quality that fits your target, rather than applying a fixed compression level. That usually keeps faces and textures looking natural.
Re-encoding also strips EXIF metadata, including GPS coordinates and camera settings. That is useful for privacy, but it means the downloaded file will not keep the original capture details.
- Works with baseline and progressive JPG files
- Corrects photo orientation before compressing
- Keeps the JPG format when you choose JPG output
- Does not add a watermark or resize by default
When to convert instead of compress
If file size matters more than universal compatibility, convert the JPG to WebP. WebP usually produces a smaller file at the same visual quality. If you need to edit the image repeatedly, keep a high-quality JPG or convert to PNG for the editing session, then export a JPG for the final upload.
Frequently asked questions
Does compressing a JPG reduce quality?
Yes, JPG is lossy. The goal is to remove as little visible detail as possible while reaching your target size. Keep the original if you may need it later.
Why did my JPG get smaller without changing dimensions?
The encoder lowers the quality setting and may also optimize how the image data is stored. Dimensions stay the same unless the target still cannot be met.
Does the tool remove EXIF data?
Yes. Re-encoding through the Canvas API removes most EXIF metadata, including GPS location. The image orientation is applied to the pixels first.
Can I compress many JPGs at once?
Yes. Add multiple JPG files, set one target size, and download the results as a ZIP when they finish.
Last updated: November 12, 2026