Image Compressor
Compress JPG, PNG, or WebP images to a smaller file size for faster uploads and page loads — pick a quality level and preview the result before downloading. Everything happens in your browser.
Your images are processed on your device and are never uploaded to our servers.
Drag images here or choose them
You can select more than one image (JPG, PNG, or WebP)
What this tool does
Shrinks image file sizes by re-encoding them at a quality level you choose. Each image is decoded and rewritten through the browser's canvas, trading a controlled amount of visual fidelity for a smaller file.
When to use it
- Site images are slowing a page down and need shrinking before upload.
- A platform is rejecting images for exceeding a size limit.
- You want to archive a photo set in less space without changing its dimensions.
- Email attachments are over the limit and you do not want to drop any images.
How to use it
- Drag your images onto the drop area or choose them — several at once, in JPG, PNG or WebP.
- Set the quality level: lower means a smaller file and more visible loss.
- Wait while they process; you can cancel or retry any individual image.
- Click "Download" per image or "Download all as ZIP".
A practical example
Input
Ten phone-camera JPGs totalling 28 MB, at a mid quality level.
Output
Ten images at the same dimensions and a far smaller total size, downloadable together as a ZIP.
The dimensions never change — this tool reduces file size, not image size. To change the pixel dimensions, use the Image Resizer.
Limitations
- The compression is lossy by nature; pushing the quality down far produces visible artefacts, especially around text and hard edges.
- An image that is already efficiently compressed may not get smaller, and the tool tells you plainly that compression did not reduce the size at that level rather than handing you a larger file.
- PNGs with transparency can lose it if you pick a format that does not support it.
- Only JPG, PNG and WebP are supported.
Privacy and processing
Every image is processed inside your browser through the canvas, and none is uploaded to our servers at any stage.
Technical notes
- Batch runs process one image at a time rather than in parallel, because re-encoding is CPU-heavy and running several at once exhausts memory on low-end devices and freezes the interface.
- That sequencing is what keeps the per-image cancel button responsive while work is under way.