About this tool
Compression is done using the browser's Canvas API to re-encode your image as JPEG at the chosen quality level. Lower quality means smaller file size but more visible artifacting. There is no automatic target-size mode: to get under a form's maximum file size, drag the slider down while watching the compressed size, which updates at every step. If even the lowest setting is still too large, shrink the image's dimensions with the Image Resizer first.
How to use the Image Compressor
- Choose an image — click the file field and select a JPG or PNG from your device.
- Drag the quality slider — the compressed preview and file size update instantly at every step.
- Compare the two sizes — the "Saved" figure shows the percentage reduction versus the original file.
- Download — click "Download compressed image" once you're happy with the size/quality trade-off.
Common use cases
- Getting a profile photo or document scan under a form's upload size limit
- Shrinking product photos so a website page loads faster
- Reducing email attachment size before sending several photos at once
- Preparing images for a portfolio or blog without a noticeable quality drop
Worked example
Photos straight off a phone camera are often several megabytes each. If an application portal caps uploads at 1MB, start the quality slider around 70–80% and check the compressed size shown under the preview — for many photos that alone gets under the limit with very little visible difference from the original. If you're still over the limit, drop the quality further; lower settings start to soften fine detail (like individual hairs or small text in the background). Watch the "Saved" percentage as you drag to find the lowest setting that still looks right to you.
Things to watch out for
Compression always re-encodes as JPEG, so a PNG with transparency will lose its transparent background (flattened to white) once compressed — use the Image Resizer or a dedicated format converter first if you need to keep transparency. Very low quality settings (below roughly 30%) introduce visible blocky artifacts, especially around sharp edges and text within a photo, so there's a point of diminishing returns where further size reduction isn't worth the visual cost. If your source file is already a low-resolution or heavily compressed JPEG, running it through again won't recover quality — compression can't add back detail that's already been discarded.
How to get a photo under 100 KB, 50 KB or 20 KB
- Load the image and note its original size.
- Drag the quality slider down until the compressed size shows under your limit, stopping at the highest quality that fits.
- If you reach 10% and it is still too big, the problem is the pixel count, not the quality. Shrink the width and height with the Image Resizer, then come back and compress again.
Check what the form means by KB. This tool counts 1 KB as 1,024 bytes, so a readout of 100 KB is 102,400 bytes. If a site defines its limit as 100,000 bytes, aim for 97.6 KB or less on the readout to be safe.
Why pixel size matters more than the slider
A file-size target is a fixed budget shared by every pixel. The table shows how many bits each pixel gets at each target size. The fewer bits per pixel, the harder the encoder has to work and the more detail it discards.
| Dimensions | Megapixels | Bits per pixel at 100 KB | At 50 KB | At 20 KB |
|---|---|---|---|---|
| 4,000 × 3,000 | 12.00 MP | 0.07 | 0.03 | 0.01 |
| 1,920 × 1,440 | 2.76 MP | 0.30 | 0.15 | 0.06 |
| 1,600 × 1,200 | 1.92 MP | 0.43 | 0.21 | 0.09 |
| 1,200 × 900 | 1.08 MP | 0.76 | 0.38 | 0.15 |
| 800 × 600 | 0.48 MP | 1.71 | 0.85 | 0.34 |
A 12-megapixel photo has less than a tenth of a bit per pixel to work with at 100 KB, while the same image at 1,200 × 900 has more than ten times as much. For small targets, resizing first usually gives a sharper result than pushing the quality slider to the bottom.
What else changes when you compress here
- The download is always a JPEG named
compressed-image.jpg, whatever format you loaded. - Photo metadata is not carried over. The image is redrawn onto a canvas and saved from its pixels, so EXIF details stored in the original, such as camera settings and GPS location, are not included in the new file.
- "Saved" never goes below 0%. If re-encoding makes the file larger, which can happen with an image that was already small or heavily compressed, the figure stays at 0%. In that case, keep your original.
- The size shown is calculated from the encoded data before download, and it matches the saved file to within a couple of bytes.
Frequently asked questions
No — compression reduces file size at the same pixel dimensions by adjusting encoding quality. Use the Image Resizer if you also want to change width or height.
70–85% typically cuts file size by more than half with minimal visible difference at normal viewing size. Below 60%, artifacts become more noticeable, especially around sharp edges and text.
You can load a PNG, but it is always re-encoded as a JPEG, so any transparent areas are flattened to white. To keep transparency, use a dedicated PNG optimizer instead.
No — compression happens locally using the Canvas API. Your image never leaves your browser.
It depends on the convention. In decimal units 0.1 MB is 100,000 bytes, but this tool uses binary units, where 1 KB is 1,024 bytes and 1 MB is 1,048,576 bytes. If a form lists its limit in MB, leave a few percent of margin under the stated number.
Not with this tool, which always saves a JPEG. The Image Converter on this site can save PNG, JPG or WebP. Check that the site you are uploading to accepts WebP before you convert, because some upload forms only allow JPG and PNG files.
At 20 KB a large photo has very few bits to spend on each pixel, so the encoder smooths away fine detail. Resize the image to smaller dimensions first, then compress it. Fewer pixels at a higher quality usually look sharper than many pixels at the lowest quality.