Image Compressor

Why images need compressing

Images are usually the single heaviest part of a web page or an email attachment. A camera photo straight off a phone commonly weighs 3 to 8 megabytes, and a page that loads several of those uncompressed will feel slow even on a fast connection, because the browser has to download every byte before it can paint the picture.

Compression exploits the fact that the human eye is much less sensitive to fine colour and brightness variation than image formats naively assume. A good compressor throws away exactly the detail nobody would notice and keeps the detail that matters, which is why a well compressed photo can look identical to the original while taking a fraction of the space.

• Compress a photo before it goes on a website, since page weight directly affects load time and search ranking.
• Compress before attaching an image to an email with a size limit.
• Compress screenshots before pasting them into documentation or a support ticket.
• Skip compression and use a resizer instead when the file is already small but the pixel dimensions are too large.
• Combine both: resize first to the display size, then compress, for the smallest possible result.

Reading the quality slider correctly

The quality setting controls how aggressively the encoder discards detail, and it is not a straight line. Above roughly 92 the file size grows quickly while the visible improvement flattens out to almost nothing. Below about 60 the eight by eight pixel blocks that JPEG style encoding works in start to become visible as soft blotches, especially around text and sharp edges.

For almost all web and document use, a quality setting between 75 and 85 is the sweet spot: the loss stays below what a casual viewer notices while the file keeps shrinking. Push higher for images that will be viewed large or zoomed into, and drop lower only when you are fighting a hard size limit and the image will be displayed small.

This tool reports the projected file size before you commit, so the most reliable method is to watch the number as you move the slider and stop where the size stops falling in a meaningful way rather than guessing a number in advance.

Lossy and lossless compression, and what happens to transparency

Lossy compression, the kind this tool performs when it re encodes as JPG, discards some original data permanently in exchange for a much smaller file. Lossless compression, which applies when the output stays PNG, rearranges the data more efficiently without discarding anything, which produces a smaller reduction but preserves every pixel exactly.

JPEG has no alpha channel, so if your source image has transparency, compressing it to JPG requires filling those transparent areas with a solid colour first. If preserving transparency matters, keep the output as PNG or WebP, both of which can be compressed while keeping an alpha channel intact.

How this image compressor works under the hood

The tool decodes your file with the browser's native image decoder, draws the pixels onto an HTML canvas, and calls the canvas toBlob method with your chosen quality factor to produce a re encoded file. That is the same primitive every image editor on the web relies on, just wired up to a slider you control directly.

Because compression happens locally, there is no upload request to intercept and no server side retention policy to trust. You can verify this yourself by opening your browser's network panel while compressing an image and watching for any request that carries the file, there will not be one. It also means there is no daily cap and no queue, since the only limit is your own device's available memory.

Compressing several images at once

For a single file, this image compressor shows you the before and after size immediately so you can fine tune the quality by eye. If you need to reduce image file size across a whole batch of photos with one setting applied to every file, the bulk image resizer handles many files sequentially and reports the saving for each one individually.

Sequential processing, rather than compressing every file at once, actually finishes faster in practice, because it avoids the memory pressure and garbage collection pauses that come from decoding several large images simultaneously.