Image compressor

Why compress images

Images are usually the heaviest thing on a web page, and oversized ones have real costs: slow page loads (which hurt both visitors and search ranking), blown email-attachment limits, and wasted storage. A photo straight off a phone can be several megabytes; the same picture, sensibly compressed, often looks identical at a fraction of the size. Compression buys that reduction by re-encoding the picture more efficiently — and, for lossy formats, by discarding fine detail the eye barely registers. The goal isn't the smallest possible file; it's the smallest file that still looks right for where it'll be used.

Lossy vs. lossless

There are two fundamentally different approaches. Lossy formats — JPEG and WebP — throw information away to save space, exploiting the fact that human vision is far more sensitive to brightness than to fine color and high-frequency texture. Lowering the quality slider tells the encoder to discard more, which is why you can shrink a photo dramatically before any difference is visible. Lossless compression — PNG — keeps every pixel exactly and saves space only by encoding repetition efficiently. That makes PNG perfect for screenshots, logos and flat-color graphics with large uniform areas, and poor for photographs, which have little exact repetition to exploit.

Choosing a format and a quality level

Match the format to the image. Photographs → JPEG or, better, WebP, which is typically 25–35% smaller at the same visual quality. Graphics, screenshots, anything with sharp edges or transparency → PNG (or WebP, which also supports transparency). For the lossy formats, the quality setting is the dial that matters: 75–85% is the usual sweet spot, lower for thumbnails and backgrounds, higher for hero images. The live preview plus the before/after sizes here let you find that point by eye rather than guessing — nudge the slider down until you can just start to see a difference, then back off one step.

Done in your browser — and what to pair it with

This tool draws your image to an HTML canvas and re-encodes it locally, so the original file is never uploaded and nothing is stored — fine for personal photos, client work or anything confidential. One tip: always compress once from your best original, since repeatedly re-saving a lossy image accumulates artifacts (generation loss) without real extra savings. If the image is also bigger than it needs to be in dimensions, resizing first saves far more than compression alone — use the image resizer — and to switch formats outright, the image converter handles that. To strip location and camera metadata, see the EXIF stripper.