Video tools

These tools all do the same underlying thing: they take the compressed video and audio out of one container and write them into another, without ever decoding a frame. That single decision is why they finish in milliseconds, why they cannot make your video look worse, and why there is no file size limit — and it is also why some things you might expect a video site to do are honestly not here.

Changing the container versus re-encoding: the distinction everything rests on

A video file has two layers. Inside are the compressed tracks — a stream of H.264 or HEVC video frames, a stream of AAC audio frames. Around them is the container: MP4, MOV, MKV, WebM. The container does not compress anything. It stores the tracks, an index of where each frame lives, timing information, rotation and language tags.

Rewrapping, or remuxing, means keeping the compressed tracks exactly as they are and writing a new container around them. Nothing is decoded, so nothing can be lost, and the work is basically reading and writing bytes. Re-encoding, or transcoding, means decoding every frame back to pixels and compressing it again. That is the only way to change resolution, bitrate or codec — and it always loses something, because these codecs are lossy by design.

The practical gap is enormous. In our measurements, rewrapping a minute of 1080p took 87 milliseconds; re-encoding the same minute in a browser took 21.6 seconds — about 250 times longer — and produced a file that is no longer bit-identical to the source. That ratio is the reason this category exists in the shape it does.

Why we have no size limit and no queue, and the sites you are used to do

Every well-known online video converter uploads your file, transcodes it on a server, and gives you a download link. That is not laziness on their part; a general converter has to re-encode, and re-encoding needs real processor time. Processor time is the bill they pay, so it becomes the thing they ration: 100 MB caps on free plans, a few conversions per hour, watermarks, a queue at busy times, and a paid tier that removes the limits they had to invent.

Rewrapping has no such bill. It happens in your browser tab using code your browser already ships, so there is nothing for us to ration and nothing to upsell. There is also no upload, which means no retention window to trust and no shareable link to leak. You can verify that part yourself: open your developer tools, watch the Network tab and run any tool here. Your file is in none of the requests, and the Content-Security-Policy header this site sends restricts connections to our own origin so a script could not send it even if it wanted to.

The honest cost of this approach: we can only give you containers your existing tracks are allowed to live in. We cannot make a video smaller, change its resolution, or turn an incompatible codec into a compatible one. Those all need an encoder.

What is not here yet, and why

There is no video compressor on this site. Compressing video means re-encoding it, and while a browser can do that, doing it well is a different project: it needs sensible bitrate decisions, a way to preview the result, and enough speed that a ten-minute video is not a twenty-minute wait. We measured the speed and the quality before deciding, and the quality is fine — a browser encode came out within a point of a desktop encoder on the standard perceptual metric. It is the workflow, not the capability, that we have not built.

There is no MP3 export either. Almost all video carries AAC audio, so extracting it without re-encoding gives you the best version that exists; making an MP3 would compress an already lossy track a second time. That is a real tool for a real need and it is coming, but it is a separate decision with a real cost, so we are not shipping it hidden inside something labelled as lossless.

And there is no frame-accurate trimming. Cutting on an arbitrary frame requires re-encoding the opening section, so our trim snaps to the nearest keyframe before your chosen point and tells you where it landed. For most trims that is invisible; on phone footage, where keyframes can be four seconds apart, it is not, which is exactly why the tool reports the real number instead of the one you typed.

Keyframes, MOV versus MP4, and HEVC — three things worth knowing

Keyframes are complete pictures; the frames between them only describe changes. This is why video compresses so well and why you cannot cut anywhere you like without re-encoding. Exported video usually has a keyframe about every second or two; phone recordings often go two to four seconds. The trim tool measures the spacing in your own file and shows it, so the precision you get is a number you can see rather than a surprise.

MOV and MP4 are unusually close relatives: MPEG built the MP4 container on the QuickTime format Apple designed, so the internal structure is nearly the same and converting between them is mostly a matter of the brand field near the start of the file. It is worth being precise about the reason people need that conversion, because the common explanation is wrong: browsers play MOV perfectly well. It is upload forms, TVs, Android phones and Windows editors that reject the extension.

HEVC, or H.265, is what an iPhone records by default because it reaches similar quality at roughly half the bitrate of H.264. The cost is compatibility: HEVC needs a licensed decoder, so plenty of desktop software cannot play it. Rewrapping HEVC works even on a machine that cannot decode it, because copying a compressed frame never decodes it — we verified that rather than assuming it. What such a machine cannot do is show you the picture, and the tools check for that before you start and say so, instead of leaving you with a black rectangle to interpret.