Why convert a JPG to AVIF
AVIF is a newer image format built on the AV1 video codec's compression, and it does a genuinely better job than JPEG at the same visual quality — typically 30–50% smaller, depending on the photo. Smooth, low-detail images compress the most dramatically; busy, high-detail or already-noisy photos see a smaller gain, since there's less redundancy for either format to exploit.
For a website, that difference is a direct page-speed win: the same photo, noticeably smaller, means faster loading with no visible quality cost.
What you lose: mostly nothing, with one caveat
At the quality settings offered here, AVIF looks the same as the JPG it came from in normal viewing. It's still a lossy format below maximum quality, same as JPEG, but it tends to hold detail better than JPEG at an equivalent file size — which is the entire point of switching.
The one real caveat is encoding time and compatibility, not image quality. AVIF takes noticeably longer to encode than JPEG or WebP, because AV1's compression is more computationally intensive. And while every current browser (Chrome, Firefox, Safari, Edge) displays AVIF natively, some older image editors, phone galleries and niche platforms haven't added support yet. If a file needs to open reliably in software you don't control, keep a JPG copy alongside it.
Choosing a quality setting
| Setting | When to use it |
|---|---|
| Maximum (95%) | Archival copies, or a source photo you'll edit further later |
| High (85%) | The sensible default for most photos - hard to distinguish from the original |
| Medium (75%) | Web images where load speed matters more than pixel-perfect detail |
| Small file (60%) | Thumbnails, previews, anywhere the image is displayed small |
Because AVIF already compresses more efficiently than JPEG, a given quality percentage tends to look better on AVIF than the same number would on a JPEG - you can often drop a step lower than you're used to and still not see a difference.
Converting a batch
Drop in several JPGs at once and each one is converted independently, then bundled into a single ZIP for download. A mix of photo sizes and subjects in the same batch is fine - there's no requirement that they be related or similar.
This tool tests whether your specific browser can actually encode AVIF before offering it - some older browsers can display AVIF but not produce it. If AVIF isn't available, the option is disabled rather than silently handing back the wrong format.
One conversion, not a chain
Convert from the original JPG rather than from a copy that's already been re-compressed elsewhere. Each lossy re-encode - JPG to something, back to JPG, then to AVIF - compounds quality loss beyond what a single conversion from the source would produce. If you have the original file, start from that.
Everything happens in your browser using the canvas API. The photo is never uploaded to a server, there's no account or sign-up, and no file size or count limit beyond what your own device can handle. As a side effect of re-encoding through canvas, EXIF metadata (camera details, GPS) does not carry over - keep the original if you need that data later.