Cut MP4 video by setting start and end times. No re-encoding preserves original quality with fast processing.
Process files in seconds with our optimized servers
Frame-accurate cuts with intuitive timeline controls
Maintain original quality with smart re-encoding
.mp4 or .m4v clip from your computer. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared.HH:MM:SS.sss (millisecond precision). The cut is frame-exact: XConvert decodes the segment and re-encodes it from your chosen timestamp at the visually-lossless default preset, so it starts precisely where you ask instead of snapping to a keyframe.MP4 (formally ISO/IEC 14496-14, an MPEG-4 Part 14 container) is the de-facto delivery format for the modern web: H.264 or H.265 video plus AAC audio in a file that streams on Chrome, Firefox, Safari, Edge, iOS, Android, smart TVs, and every social platform. "Cutting" an MP4 means extracting a sub-range of the timeline — often to keep a highlight, remove dead air, or split a long recording into shareable pieces.
Editing vocabulary overlaps, and most online tools blur the distinction. Here's what each verb usually means and which xconvert tool maps to it.
| Verb | What It Does | xconvert Tool |
|---|---|---|
| Trim | Shorten by removing the head, the tail, or both — keeps one continuous middle section | Trim MP4 or Video Trimmer |
| Cut | Extract a sub-range by start time + duration (or remove a middle section) | This page / Video Cutter |
| Split | Divide a long file into multiple independent output clips | Video Cutter (re-run per segment) |
| Merge | Join multiple clips end-to-end into a single MP4 | Upload multiple files to the relevant convert tool |
For a single-segment extract using start time + duration this page is correct. For multi-segment workflows (e.g. extracting three highlights from one source) re-run the cut once per segment — each run produces an independent MP4 you can later combine.
MP4 video is compressed as a sequence of keyframes (I-frames, fully self-contained) and predicted frames (P/B-frames, which reference earlier frames). A cut can only start on a keyframe without decoding — which is why stream-copy cutters snap away from your mark. XConvert instead decodes the segment and re-encodes it at the visually-lossless default preset, so every cut is frame-accurate.
| Mode | Output Start | Quality | Speed | Notes |
|---|---|---|---|---|
| Stream-copy (other tools) | Snaps to nearest preceding keyframe (commonly 1-10 s before your timestamp) | No recompression, but the start point is off your mark | Very fast | Not how this page cuts — some desktop cutters work this way |
| Re-encode (how XConvert cuts) | Frame-accurate to your chosen timestamp | One visually-lossless-preset pass — visually indistinguishable at the defaults | CPU work proportional to clip length | Every cut on this page, including removing a 0.3 s blooper or aligning to subtitle cues |
Modern social-media recorders (OBS, ShadowPlay, iPhone screen capture, GoPro) typically place keyframes every 1-2 seconds, while long-GOP broadcast captures and older AVCHD camcorder files stretch that to 5-10 seconds — but because XConvert re-encodes the kept range, none of that affects where your cut starts.
Not visibly. The cut decodes the selected range and re-encodes it at the visually-lossless default preset (CRF 18 for H.264 / CRF 17 for HEVC). That is one lossy generation on paper, but at the default settings it is visually indistinguishable from the source — and it is what makes the cut frame-accurate instead of keyframe-snapped.
With XConvert it shouldn't — cuts here are re-encoded, so the output starts at exactly the timestamp you typed. The snap-back you're describing is how stream-copy cutters behave: they can only begin output on a keyframe, so a requested 00:01:13 cut ends up starting at 00:01:11 if that's the nearest preceding keyframe.
Yes — re-run the cut once per segment, varying Start time and Duration. Each run downloads as an independent MP4. To stitch the resulting clips back together into one continuous file, drop them into a video-merge workflow. There's no "multi-range" export in a single pass on this page.
Cuts run on our servers, so the practical ceiling depends on upload size and connection speed and the source bitrate. Multi-GB 4K H.265 files generally work on desktops with 8 GB+ free memory; phones and Chromebooks should stick to smaller clips. There is no hard server-side cap.
The output keeps the source resolution and frame rate, and stays MP4 with a standard codec — but the video is re-encoded at the visually-lossless default preset, so the exact bitrate will differ from the source even though the picture looks the same. Override codec or size by selecting a different codec or changing Video resolution before cutting.
If the source plays everywhere, the cut will too — the output is a standard MP4 with H.264/H.265 video and AAC audio. MP4s written by this tool include a faststart-friendly atom layout, so they begin playing on the web before the full file downloads. For older devices that struggle with H.265, convert the cut output to H.264 with MP4 to MP4 compression.
The trim tool is structured around "remove head and/or tail" — you mark in/out points on a continuous segment. This cut tool is structured around "extract a segment of length N starting at time T". Functionally they overlap for the simple case (one middle segment kept). Pick whichever workflow matches how you think about the edit.
This page outputs an MP4 with both streams. To strip the audio into its own file, use MP4 to MP3 or the standalone audio cutter. Both run on our servers.
The accepted file types here are .mp4 and .m4v only. For other containers use the format-specific cutters: Cut MOV, Cut MKV, or Cut WebM. The processing model is identical — only the container changes.