More free video tools
Same quality bar. Still free. Still no upload.
Your files never leave your device
Free to use, with no account and no watermark. The first run downloads the model files once and your browser caches them; the model then runs on your device, and the file you process is never sent anywhere.
This is not a promise about how carefully we handle your upload. There is no upload. The file you choose is read into memory by your own browser, processed there, and handed back as a download — the tool has no server component and makes no network request with your data. The only download is the model itself, which contains nothing of yours.
- Nothing is stored. We never receive the file, so there is no copy to retain, no retention period to disclose, and nothing to delete on request.
- Nothing is transmitted. No upload endpoint, no analytics payload carrying file contents, no third-party processor.
- You can verify it. Run the tool once so the model is cached, then disconnect from the internet and run it again. It still works, because everything it needs is already on your machine.
- Nothing persists after you close the tab. Files are held in page memory only — not in local storage, not in a cache, not in a database.
A note on wording, because it matters: we do not describe these tools as “encrypted.” Encryption protects data that travels to somebody else's computer. Your file does not travel, so there is nothing to encrypt and nothing to intercept — which is a stronger guarantee than encryption, not a weaker one. The page itself is served over HTTPS like the rest of the site.
This makes the tools safe for material you are contractually barred from uploading to third-party services — client footage under NDA, unreleased campaign assets, anything covered by a confidentiality clause.
How it works
1. Drop a landscape video
MP4, MOV or WebM up to 2 minutes. It stays in this tab. The first run downloads the 15 MB detection model once into your browser cache.
2. Track the subject
D-FINE nano checks two to four frames a second on your device with WebAssembly, so it works in any modern browser. People come first; hard cuts are found so the crop switches on the edit instead of panning across it.
3. Check and adjust the crop
Scrub the preview to see the window the export keeps. Pick 9:16, 4:5 or 1:1, Steady, Balanced or Follow movement, and biggest subject or lock on one. Click a person to lock onto them. Changes apply instantly, with no re-tracking.
4. Export with the original sound
The clip plays once in real time and is recorded at the crop's native resolution with the source audio, usually as WebM. Download it for TikTok, Reels or Shorts.
Questions
How does the auto reframe decide what to follow?
An object detector looks at two to four frames a second and finds people and 79 other everyday object types. By default the crop follows whoever is largest in frame, people first, aiming at the upper body so faces stay in shot. Switch to Lock on one to stay with the first subject, or click the person you want in the preview. The tool does not recognise faces or identities, only where a person is.
Is my video uploaded anywhere?
No. Frames are read, analysed and re-encoded inside this browser tab. The only network request is the first-run download of the open detection model from Hugging Face, which your browser then caches.
What are the limits?
Clips up to 2 minutes, because export plays the video once in real time while it records. Output is the crop at your source's native resolution, so a 1080p landscape clip gives a 608 x 1080 vertical video; it is not upscaled. Tracking takes roughly a quarter to the full length of the clip, depending on your device. Safari usually cannot record a canvas stream, so export needs Chrome, Edge or Firefox.
Why does the crop not move when my subject shifts slightly?
On purpose. The crop only moves once the subject leaves a band around the centre, then glides there, which is what stops the wobble most auto-crops have. Steady uses the widest band, Follow the narrowest. The path is smoothed without overshoot and never leaves the frame.
Which model does it use, and can I use the output commercially?
D-FINE nano (onnx-community/dfine-nano-coco-ONNX), released under the Apache-2.0 licence, which allows commercial use. The video you export is yours; the model only decides where the crop sits.
Is the export lossless, and which format is it?
No. The browser redraws each frame at the new shape and MediaRecorder encodes a new file, usually WebM (MP4 in some browsers), with the original audio. Quality lands close to the source. TikTok, Instagram and YouTube accept WebM uploads.
Related tools
Stay in the same job cluster. Free tools stay on-device; paid tools are labelled as such.
These rearrange files. Versely makes new ones.
Everything on this page works on a file you already have, which is why it costs nothing. Generating something that did not exist (video from a prompt, a voice, a score) runs a model, and that costs credits.