JPEG is lossy compression tuned for human vision, and no alpha channel at all. WebP is lossy or lossless depending on the encoder, with alpha and animation. ImageMagick sits between the two, and this page says exactly what happens to a JPG file on the way to becoming a WEBP one.
What runs when a JPG file becomes WEBP
ImageMagick 6.9.12, on a machine we rent and watch. Your JPG file is uploaded once, ImageMagick runs once, the WEBP comes back, and neither file is kept. JPG to WEBP is one of the 3649 pairs that engine was probed on with a real file, which is why it has a page here and why the pairs the probe could not prove do not.
What survives from the JPG into the WEBP
The pixel dimensions of your JPG file. The WEBP comes out the size the JPG went in, because nothing on this path resizes anything.
What JPG to WEBP costs you
Nothing measurable, which is rare enough to say plainly. JPG and WEBP carry the same kind of content with the same fidelity, so this pair changes the container and not the picture. What does change is the size, and often by a lot.
If a model is going to read the WEBP
A WEBP file is pixels, and pixels are expensive for a model to read. Handed this WEBP, an assistant spends its budget looking at the picture rather than reading it, and it will paraphrase what it thinks it saw. If your JPG is mostly words, the useful move is to pull the text out of it, not to change which picture format holds it.
JPG to WEBP, measured rather than promised
Running JPG to WEBP against the real engine with a real file, ImageMagick wrote 350 bytes of WEBP in under a tenth of a second, machine otherwise idle. That is one JPG file on one day and not an average, which is why the number is given together with the file that produced it.
The JPG to WEBP verdict was reached by a format signature, meaning at least four bytes unique to the format were found at the right offset. Which check was used matters, because they do not all prove the same thing, and a status code of 200 proves nothing whatsoever about whether the WEBP file has anything inside it.
The same JPG file, sent somewhere else
Other outputs the probe measured out of a JPG file, so the cost of choosing WEBP can be read against something. Sizes do not compare across engines, because each family was probed with its own JPG witness file.
JPG to PDF: 4,232 bytes in under a tenth of a second, verified by a format signature.
JPG to PNG: 805 bytes in under a tenth of a second, verified by a format signature.
JPG to GIF: 3,487 bytes in 0.1 seconds, verified by a format signature.
JPG to TIFF: 72,264 bytes in under a tenth of a second, verified by a format signature.
Other ways into WEBP, and what they measured
Among the published routes into WEBP, JPG is the eighth largest output of the 8 measured. The witness files differ, so this ranks the probe run and not your document.
PDF to WEBP: 906 bytes in 0.1 seconds.
HEIC to WEBP: 416 bytes in under a tenth of a second.
HEIF to WEBP: 416 bytes in under a tenth of a second.
ICO to WEBP: 370 bytes in under a tenth of a second.
SVG to WEBP: 608 bytes in under a tenth of a second.
What we will not pretend about JPG to WEBP
A JPG file over 25 MB is refused before the upload finishes rather than after it, so you do not wait for a rejection.
A JPG to WEBP run that passes 30 seconds is killed, and the ImageMagick process is killed with it. A run left behind would sit on one of the machine's two cores until somebody noticed.
No text is read out of the JPG anywhere on this path. Converting a picture to WEBP never turns it into words, whatever the picture happens to contain.
ImageMagick was not saturated at sixteen simultaneous conversions: it held 47.5 a second while latency moved from 0.22 s to 0.34 s, and the service memory did not move off 51 MB. That is the measured headroom behind JPG to WEBP, not a marketing number.
One JPG file at a time, chosen in the browser. There is nothing else to set up and nothing else on offer.
JPG to WEBP: what people ask
What actually converts my JPG file to WEBP?
ImageMagick does it, 6.9.12, on a machine we rent and watch. Not a browser trick and not somebody else service: the JPG file is uploaded once, ImageMagick runs once, the WEBP comes back, and neither file is kept afterwards.
How long does JPG to WEBP take?
On the file the probe used, ImageMagick took under a tenth of a second and wrote 350 bytes of WEBP. That is one real measurement on one real JPG file, not an average and not a promise about yours: a larger JPG takes longer, and past 30 seconds the run is stopped.
What do I lose going from JPG to WEBP?
The one to know about first: Nothing measurable, which is rare enough to say plainly. JPG and WEBP carry the same kind of content with the same fidelity, so this pair changes the container and not the picture. What does change is the size, and often by a lot.
Does the WEBP come out the same size as the JPG?
The same pixel size, yes: nothing on this path resizes anything, so the WEBP is as wide and as tall as the JPG was. The number of bytes is a different question and usually moves a long way, which is normally the reason for doing this at all.
Is JPG to WEBP free?
There is a free allowance every month, and one JPG to WEBP conversion costs half a credit against it. When the allowance runs out the tool says so and stops, rather than quietly handing you a worse WEBP.