Image Compressor
Reduce image file sizes before publishing them on a website, sending them by email, or adding them to a product listing.
- Start with a moderate quality setting, then lower it only if the file is still too large.
- Prefer WebP when the destination supports it and you need smaller files.
- Resize oversized images to their display dimensions before aggressive compression.
Supports PNG, JPG, JPEG, and WebP. Processing happens entirely in your browser, so your image files are never uploaded or stored.
Why compress images
Large images slow down pages, burn mobile data, and push Largest Contentful Paint later than it needs to be. Compression reduces bytes while keeping the picture usable for the web, email, or documentation.
Compression is not the same as resizing. If an asset is 4000px wide but displays at 800px, resize first, then compress. Shipping unused pixels wastes more than a slightly lower quality setting.
A sensible compression workflow
Use this order when preparing an image for the web:
- Keep a lossless original somewhere safe.
- Crop or resize to the real display size.
- Convert to a suitable format if needed (often WebP or JPEG for photos).
- Compress until the file is small enough without obvious artifacts.
- Strip metadata if the image will be shared publicly.
Private, local processing
Screenshots and product photos can include sensitive UI, customer data, or unpublished work. This compressor processes files in your browser — they are not uploaded to a remote encoding service.
Lossy, lossless, and generation loss
JPEG and WebP below quality 100 are lossy: they discard visual information the eye is unlikely to miss. PNG is lossless — every pixel survives exactly — which is why a PNG of a photograph is many times larger than a JPEG of the same image. Compression here re-encodes the pixels, so it can only preserve what the format allows.
The trap is generation loss. Re-compressing a file that is already lossy throws away more information every time, and the damage compounds: a JPEG saved at 60%, then saved again at 60%, looks worse than the same file saved once at the equivalent lower setting. Always compress from the original export or camera file, never from a copy you already optimized.
What the quality setting actually controls
Quality is a quantization knob, not a percentage of goodness. JPEG and lossy WebP use it to decide how aggressively to discard high-frequency detail — the part of the image your eye notices least. Turning it down shrinks the file fast at first, then stalls.
The useful range for photographs is usually 70–85. Below roughly 60, blocky artifacts appear around edges and in smooth gradients, especially skies and skin. Because the curve flattens, the last few points of quality cost a lot of bytes for little visible gain: set the highest quality that still meets your size budget, and judge it at the size the image will actually appear, not at 100% zoom.
Frequently asked questions
Will compression reduce image quality?
Lossy formats (JPEG, WebP at lower quality) trade some fidelity for size. Start higher and reduce until the file meets your limit. PNG compression is typically lossless for the pixel data but still benefits from removing unnecessary complexity.
Should I compress or resize first?
Resize to the display dimensions first when the source is far larger than needed. Then compress. Compressing a huge unused resolution still leaves you with an oversized asset.
Are my images uploaded?
No. Encoding runs locally in your browser. The file stays on your device.
Does compressing an image more than once make it worse?
Yes, when the format is lossy. Each re-encode discards more information, so quality drops a little every pass. Compress from the original each time rather than re-compressing an already-optimized file.
What quality setting should I use for photos?
70–85 is the usual range. Start at 80, check edges and gradients at the size the image will display, and lower it only until the file meets your limit. Below about 60 the artifacts become easy to see.