convert to
resize
✓ EXIF, GPS and all other metadata are removed from the output.
Drop images here
or click to browse — PNG, JPEG, WebP, GIF, HEIC, AVIF, SVG — you can also paste images (Ctrl+V) anywhere on this page
Turn an AVIF into the one still-image format nothing refuses — and decide yourself what replaces the transparency.
Your files never leave your browser — all processing is local.
✓ EXIF, GPS and all other metadata are removed from the output.
Drop images here
or click to browse — PNG, JPEG, WebP, GIF, HEIC, AVIF, SVG — you can also paste images (Ctrl+V) anywhere on this page
Transparency has nowhere to go, because the target has no alpha channel at all. Every transparent and semi-transparent pixel is composited onto the background color in the settings above — white unless you change it — and the converter raises a warning the moment it finds any alpha to flatten. A cut-out product shot becomes a rectangle, and the flattened pixels are indistinguishable afterwards from pixels that were always that color, so no editor can undo it. Semi-transparent areas such as the soft edge of a shadow are blended rather than clipped, which means the edge reads correctly against that one background and wrong against every other. Pick the color the image will actually sit on, or take the PNG route, which keeps the channel intact.
This is also a second lossy generation, and the two generations leave artifacts of different shapes. The first encoder tends to smooth texture and soften fine detail; the second adds 8 × 8 blocking, ringing around high-contrast edges, and color stored at half resolution in both directions. The second set lands on top of the first rather than replacing it. That last point about color matters most for graphics: a thin colored rule or a line of red text that came through the source encode cleanly can pick up a fringe here that raising the quality does not remove, because the chroma is discarded before the quality setting's quantization step even runs.
The remaining losses are quick to list. A multi-frame source contributes only its first frame, with a warning. The decoded buffer is 8-bit, so any 10- or 12-bit tonal range in the original is flattened before encoding and the extra levels cannot be restored. The embedded ICC profile and any metadata are not carried into the output, so the result is untagged and a viewer will read it as sRGB.
On 2026-09-12 the script measured a 1280 × 854 source at quality 85, 491,861 bytes, and encoded it to a 4:2:0 JPG at the same quality 85: 348,599 bytes — about 29% smaller. That looks like the older format winning, and it is not. The two quality scales are not a shared unit: the source was written at a higher-fidelity operating point than the number suggests, and the second encoder was asked for less. Encode both from an uncompressed master at matched *visual* quality and the comparison reverses decisively. Note also that the measurement was made with sharp, while your browser encodes through canvas with its own quantization tables, so expect a few percent either way rather than this exact figure.
Do not convert anything with transparency you care about: a logo, a badge, a sticker, or an image that has to sit on backgrounds you do not control. Do not convert an animation and expect the motion to survive. Do not convert a working master you will edit and re-save, because each pass is another generation. Do not make it a routine step for the web, where every current browser reads the source directly and this costs both quality and, at matched visual quality, bytes. It is the right call when one specific piece of software is the blocker — a photo lab or print service, an older desktop editor or office suite, a legacy publishing system, a camera or e-reader that indexes one extension, or a form whose accepted list stops at two.
Conversion happens entirely in your browser — no upload, no queue, no watermark. Drop in as many files as you like and download them individually or as a zip.
It is blended into the background color shown in the settings, white by default, and a warning tells you it happened. The result is ordinary opaque pixels that no editor can separate again, so choose the color the image will sit on — or convert to PNG, which keeps the channel.
Because the two quality scales are not comparable. At the same number the source sat at a higher-fidelity operating point than the output, so the measured 491,861 bytes fell to 348,599. It is not evidence that the older codec is more efficient — from a master at matched visual quality it clearly is not.
On a photograph at a high quality setting, little that you would notice. On flat graphics and text it can be obvious, because color is stored at half resolution and thin colored edges pick up a fringe no quality setting removes. Convert from the original rather than from an already re-saved copy.
No. Conversion happens inside your browser using the canvas API and a Web Worker — files never leave your device. There is no upload step, no queue and no file size limit beyond your machine's memory.
No — and that's a feature. Re-encoding through the canvas strips all metadata: GPS location, camera model, timestamps and color profiles. The output contains pixels only, which is usually what you want before sharing an image publicly.
Yes. Drop or paste any number of files; each is converted in the background with its own progress and download button, and you can grab everything at once as a .zip archive.
It depends on the target. The converter always decodes to raw pixels and re-encodes, whichever formats are involved, which is also how metadata gets stripped. For lossless targets (PNG, and GIF within its 256-color limit) the pixels stay identical to what was decoded; for JPEG, WebP and AVIF the quality slider controls the trade-off, and converting a file that is already lossy adds one more generation of loss.