Trim a Video Without Re-encoding
- No Sign-up
- 100% Free
- Zero Server Data
Choose a start and an end, and get that piece as a new file. Nothing is decoded and nothing is compressed again: the frames inside the cut are copied out of the original byte for byte, so a two-minute clip out of an hour takes about as long as reading the file, and the picture is exactly the picture the camera produced. The one thing to know is that a cut can only begin at a keyframe, so the start moves back to the nearest one. The page tells you how far before you press anything, and shows where those points are.
1 · The video
2 · The piece to keep
The marks under the first slider are the points a cut can start from. Play the video and use the buttons below to take the current moment.
Frequently asked questions
Why did my cut start earlier than I asked?
Because a video frame is usually stored as the difference from the frames around it, and means nothing on its own. Only a keyframe stands alone, and they are seconds apart. A cut that began anywhere else would open with grey or with smeared blocks, so the start moves back to the nearest keyframe. The page shows where they are and says how far it moved.
Could you not re-encode just the beginning?
Some tools do, and it is a real trade rather than an oversight. Re-encoding even a second means decoding and compressing again, which is slower by orders of magnitude and loses a generation of quality at the join. For a tool whose promise is that the picture is untouched, taking a little more material is the better answer.
Does the quality drop at all?
No. Not a single frame is decoded. The samples inside the cut are copied out of your file and written into a new one with a fresh index, which is why it is done in seconds rather than minutes.
Is my video uploaded?
No, and for a video that matters more than for most files. It is read in this tab. The site’s security policy sets connect-src ‘none’, so the browser blocks network requests from these pages, so it could not be sent even if the code tried, and a two-gigabyte upload is not happening in the background either.
My screen recording says it has one keyframe.
That is how the recorder wrote it: some of them put a single keyframe at the start and store the rest of the file as differences, because a screen mostly does not change. Any cut of such a file has to begin at the beginning. Nothing here can change that without re-encoding the whole thing.
What about MOV files from an iPhone?
They usually work, because a MOV and an MP4 are the same structure with a different label. A file that is not readable is refused by name rather than half-cut.