- Why does WebP or AVIF beat just lowering my JPEG quality?
- Because they compress differently, not just harder. JPEG breaks an image into 8×8 blocks and compresses each block largely in isolation, so when you push the quality down you get the classic blocky, ringing artifacts around edges. WebP (from the VP8 video codec) and AVIF (from AV1) add intra-frame prediction: before storing a block, the encoder guesses it from the already-decoded pixels above and to the left, then stores only the error. A clear sky or a studio backdrop is almost perfectly predictable, so it costs close to nothing. In practice WebP runs about 25-35% smaller than JPEG at matched visual quality, and AVIF often 20-50% smaller than WebP again, with the gap widest on flat colour and smooth gradients and narrowest on noisy, highly detailed photos.
- Should I pick WebP or AVIF?
- Pick WebP unless you have a specific reason not to. Every browser in current use displays it, every major CMS accepts it, and it encodes fast. Pick AVIF when file size matters more than anything and you know your audience is on recent browsers — AVIF display support arrived in Chrome 85, Firefox 93, and Safari 16.4, so it's broad now but not universal, and older devices will show nothing rather than a fallback. AVIF also handles 10- and 12-bit colour and HDR, which WebP can't, so it's the better archive of a wide-gamut photo. If you're serving a website, the robust answer is both: an AVIF source with a WebP fallback in a picture element.
- The AVIF option is switched off. Why?
- Because your browser can read AVIF but can't write it. Compressing in a browser tab means using the browser's own encoder through the canvas element, and while every current browser has a WebP encoder built in, none of them ship an AVIF encoder there yet — decoding and encoding are separate pieces of code, and only the decoder got added. The reason this matters: when you ask a canvas for a format it doesn't know, it silently hands back a PNG instead of an error. A tool that didn't check would give you a PNG wearing a .avif name, usually larger than the file you started with. So this one tests the encoder before offering the option, and says so when it isn't there. If a browser ships the encoder later, the option appears on its own.
- Does compressing strip my EXIF data and GPS location?
- Yes. The image is decoded to raw pixels, drawn onto a canvas, and re-encoded, and that process keeps nothing but the pixels. Camera model, lens, shutter speed, the date taken, GPS coordinates, and any embedded ICC colour profile are all gone from the output, and colours are written as standard sRGB. That's usually what you want before posting a photo publicly, since location data in a holiday snap is a real privacy leak. It's not what you want if you're archiving originals or handing files to a photo editor who needs the capture settings — keep your originals in that case and treat these as web copies.
- What does the quality number actually mean?
- It's a dial on how much detail the encoder is allowed to throw away, not a percentage of anything you can measure. It isn't comparable between formats either: AVIF at 50 and WebP at 50 are different amounts of loss, and neither matches JPEG at 50. The numbers worth knowing are the shape of the curve. Between 90 and 100 you pay a lot of bytes for differences nobody can see. Between about 60 and 90 is where most photos live — 75 is the usual sweet spot. Below 60, flat areas start to band and text edges and sharp lines go mushy, which is fine for a thumbnail and bad for a screenshot of anything with words in it. Move the slider and re-run; the before → after line tells you what you bought.
- Why did one of my files come out no smaller?
- Three common reasons. The file was already compressed hard, so there's little left to remove and the format switch can't make up the difference. The image is tiny — under a few kilobytes, the format's own header and metadata are a meaningful share of the file, and a new container can cost more than it saves. Or the source is a PNG of flat-colour line art, which PNG's lossless compression is genuinely excellent at; a lossy format has to spend bits describing edges PNG stored almost for nothing. When this happens the tool keeps your original rather than handing you the bigger version, because a compressor that makes files larger isn't doing its job.
- Are my images uploaded anywhere?
- No. Your browser reads each file from your device, decodes it, re-encodes it in memory, and gives you a download link. There's no server step and no account, which is what makes it safe for client work, medical or legal scans, and photos of your family. Close the tab and the results are gone. It's also why the limits are what they are: 10 MB per image and 200 MB per batch is roughly what a phone or an older laptop can hold in memory at once without the tab being killed.
- How is this different from the other image tools here?
- Image Compressor keeps your format — a PNG comes out a PNG, a JPG comes out a JPG — which is what you want when a system on the other end only accepts one format. This tool changes the format on purpose, because that's where the large savings are. Image Format Converter moves between PNG, JPG, and WebP without a quality dial, for when you just need a different extension. Image Resizer changes pixel dimensions, which is the other big lever on file size and often worth doing first: a 4,000-pixel-wide photo displayed in a 800-pixel column is wasting most of its bytes no matter how well it's compressed.
- What are the limits?
- Up to 100 images per batch, 10 MB per image, and 200 MB in total. Images can be up to 16,383 pixels on a side — that's not our rule, it's the ceiling built into the WebP format, which stores width and height in 14 bits. Anything larger has to be resized first. Transparency survives in both WebP and AVIF, so a logo with a clear background stays clear. Animated GIFs and animated WebPs keep only their first frame; for those, use the GIF Compressor instead.
