Image Format Converter

The formula

Natural dimensions

The image’s real pixel width and height. Nothing is resized — the output has the same dimensions as the input.

Quality

A value between 0 and 1 for lossy formats. Higher keeps more detail and produces a larger file.

Format support

Which output formats work depends on your browser. If a format is unavailable the conversion reports a failure rather than producing a broken file.

Worked example

Converting a 2,400 by 1,600 PNG screenshot to JPEG at quality 0.85.

• The PNG is decoded and drawn to a 2,400 × 1,600 canvas
• The canvas is re-encoded as JPEG at quality 0.85
• Dimensions are unchanged; the file size falls because JPEG discards detail
• Any transparent areas are filled, since JPEG has no alpha channel

A JPEG at the original 2,400 × 1,600, substantially smaller than the PNG, with transparency lost.

Transparency does not survive a conversion to JPEG

JPEG has no alpha channel, so transparent regions are filled during encoding — commonly with black or white depending on the browser. Logos, icons and anything cut out against a transparent background should go to PNG or WebP instead. This is the conversion that most often produces a surprising result, and it is a property of the format rather than of the tool.

Re-encoding a lossy image loses more each time

Converting a JPEG to another JPEG discards detail again on top of what was already discarded, and repeating that cycle visibly degrades the image. Always convert from the highest-quality original you have rather than from a file that has already been through compression. Going from JPEG to PNG does not recover anything either — it just stores the existing artefacts losslessly, usually in a larger file.

Metadata is stripped

Because the image is redrawn onto a canvas, everything that was not pixel data is discarded: EXIF, colour profile, orientation flags, GPS coordinates, camera settings. That is useful when you are stripping location data from a photo before sharing it, and a problem when you needed the profile — a wide-gamut image can shift in colour once the profile is gone. Orientation is applied by the decoder before drawing, so rotated photos come out the right way up.

Choosing a format

PNG for anything needing transparency or sharp edges and flat colour, such as screenshots, logos and diagrams. JPEG for photographs, where its compression is designed for continuous tone. WebP for either, at a smaller size than both, with support now broad enough for general web use. AVIF is smaller again but encoding support varies by browser, and this tool will tell you if yours cannot produce it.

Common questions

Does converting change the image dimensions?

No. The canvas is set to the image’s natural width and height, so the output has exactly the same pixel dimensions as the input. Only the encoding changes, which is why the file size can change dramatically while the image does not.

What quality setting should I use?

The tool defaults to 0.92, which is high. Around 0.80 to 0.85 is a common balance for photographs on the web, where the size saving is substantial and the difference is hard to see. Below about 0.7 compression artefacts become visible on detailed images.

Why did my transparent background turn solid?

Because you converted to JPEG, which cannot store transparency. Use PNG or WebP for images with transparent areas.

Are my images uploaded to a server?

No. The file is read, decoded and re-encoded entirely in your browser, and the download is generated locally. Nothing is sent anywhere, which also means the conversion works with no network connection once the page is loaded.