Compress Image
Reduce the size of a JPG, PNG or WebP by re-encoding it at a quality you pick, with the real before and after numbers shown for every file. Use it when a form rejects an oversized upload, an email refuses to send, or one 4 MB picture is making a page feel slow.
Lower means a smaller file and more visible artefacts. Around 75–80 is indistinguishable from the original for most photographs.
Keeping the original format is the safest default. WebP is usually 25–35% smaller than JPG for the same perceived quality.
How compress image works
- Add your imagesDrop in one or several JPG, PNG or WebP files. Each is checked against its real file signature before it is accepted.
- Choose a qualityStart at 75%. Drop towards 50% for maximum savings on a photo gallery, or raise it towards 90% for screenshots with fine text.
- Download the smaller filesFurtu re-encodes each image in your browser and shows the original size, the new size and the percentage saved.
What you get
Make a photo or screenshot smaller while keeping it sharp enough to use. Everything happens inside this page: the file is read by your browser, transformed in memory and handed straight back to you as a download. There is no upload queue, no waiting for a server, and nothing left behind when you close the tab.
Supported formats
Furtu accepts .jpg, .jpeg, .png, .webp, .avif, .gif, .bmp, .svg, .ico, .heic, .heif files up to 100 MB each, and processes up to 20 files in one run.
- Every file is checked against its real file signature, not just its name, so a renamed or corrupt file is rejected with an explanation instead of failing halfway through.
- Files over the limit are refused up front rather than after a long wait.
Limitations, stated up front
- Compression is a re-encode, so the pixels are decoded and written again. That is what removes the savings opportunity: a file that has already been compressed hard cannot be recovered by compressing it again.
- Transparency is lost when you convert a transparent image to JPG, because JPG has no alpha channel. The transparent areas are filled with white.
- GIFs are rejected, animated or not. A canvas round-trip cannot preserve animation, and a still GIF would come out as a silently flattened first frame, so Furtu refuses it rather than handing you something that only looks correct.
- SVG is a vector format and cannot be re-encoded through a canvas without rasterising it, so vector files are rejected with an explanation rather than returned as a lossy approximation.
Frequently asked questions
Are my images uploaded?
No. The image is decoded and re-encoded inside your browser using the codecs already built into it. Nothing is sent anywhere, which matters most for the photos people compress: passports, receipts, client work and anything shot at home.
How much smaller will it get?
A typical photograph re-encoded at 75% lands around a third of its original size. Screenshots with large flat areas compress even better. Images saved at maximum quality, or PNG files full of transparency, often shrink much less. Furtu shows the real before and after size for every file so you can see what actually happened.
Will the quality loss be visible?
At 75–85% on a photograph, almost never. Artefacts show up first in skies, gradients and fine text, and they become obvious below about 50%. If the file is a screenshot with sharp text, keep it at 90% or leave it as PNG.
Does compressing an already-compressed image help?
It helps, but with diminishing returns. Each re-encode works from the pixels already present, so a second pass can never recover detail the first one discarded. Compressing a file that was saved at quality 100 usually gains a lot; compressing one saved at quality 60 usually gains a little.
Can I compress a whole folder at once?
Yes. Add up to 20 images at once and Furtu works through them, reporting the result for each one. A file that fails is reported on its own rather than stopping the rest of the batch.