← Video Tools

Compress Video for Smaller Files and Better Balance

What compressing a video actually changes

Make a video smaller with a preset, a quality value or a target size in MB. Resolution, frame rate and audio bitrate offer separate ways to reduce the data. The result depends on the source: a file that was already compressed efficiently can grow when encoded again. Compare moving details and sound before sharing the result.

Cost: Free to use. File size and usage limits apply.

How to compress a video

  1. Upload one video

    Click Upload a video to compress or drop an MP4, MOV, MKV, WebM, AVI, M4V, MPEG or OGV file up to 500 MB (1 GB on a paid plan). One video is compressed per job.

  2. Pick a preset

    Balanced is selected initially. Presets set CRF, quality, encoder speed, audio bitrate and audio removal; you can change them afterwards. Output defaults to Same as source, or choose MP4, WebM, MKV, MOV or AVI.

  3. Set a size goal if you have one

    Switch Rate control to Target size and enter the limit under Target MB when a form, email or chat app rejects larger files. The initial target is 20 MB.

  4. Adjust resolution or codec only when needed

    Open Advanced encoder to lower width, height or frame rate, or change the codecs. Empty dimension and FPS fields keep those source properties.

  5. Compress, compare, download

    Click Compress Video!, compare file size, moving details and sound, then choose Download Compressed Video. Continue editing keeps your settings; Reset restores defaults. The result also stays in Uploads for other tools.

Which preset to choose

Each preset is a starting point with fixed values. Small and Tiny combine a higher CRF and a lower audio bitrate with the slower encoder speeds slower and veryslow, which squeeze out extra savings at the cost of encoding time.

Balanced — CRF 28, 128 kbps audio

A starting point using the slow encoder speed. Check the result: footage that was already heavily compressed can become larger.

Small — CRF 32, 96 kbps audio

For sharing links and previews where a softer image is acceptable. Fine detail such as grass or confetti is the first thing to suffer.

Tiny — CRF 36, 64 kbps audio

For hard limits and drafts. Expect blocky gradients and thin speech audio; lowering the resolution often looks better than going below this.

Social — CRF 30, fast encoding

Uses fast and 128 kbps audio. Choose it when encoding time matters more than the extra compression from a slower preset.

Archive — CRF 22, 192 kbps audio

Uses medium and keeps more detail for later editing. Another encode still costs quality; this is not a lossless archive format.

Mute — CRF 29, no audio track

Uses fast and removes the soundtrack. Useful for background loops and silent product clips.

CRF, bitrate, target size or lossless?

CRF is selected initially. The percentage slider runs from 1–100%, starting at 75%; higher values target more quality. An explicitly chosen CRF takes priority. Bitrate and target size set a data budget instead.

CRF (constant quality)
Lower values keep more detail and produce larger files; size follows the picture content. The initial value is 28. Ranges: 0–51 for H.264/H.265, 0–63 for VP9/VP8/AV1. Raising H.264 CRF by 6 roughly halves the bitrate as a rule of thumb.
BITRATE
Set Video kbps from 50–200000 kbps. Maxrate caps peaks within the same range, with Bufsize from 50–200000 kbit of buffer capacity. These fields start empty.
Target size
The tool subtracts audio from the MB budget, divides the remainder by duration and encodes in two passes. It refuses a budget below 50 kbps for the picture. Leave headroom below a hard upload limit.
Lossless
Uses the encoder's lossless mode to avoid additional picture-compression loss. Files can grow substantially. Changing resolution, frame rate or pixel format still changes the picture; AAC or MP3 audio is still encoded with loss.

Advanced encoder settings explained

Advanced encoder starts collapsed. Empty dimension, FPS and audio-format fields keep that source property; codecs, pixel format and metadata removal have their own defaults.

Video codec
Initially H.264 (libx264). Alternatives: H.265 (libx265), VP9 (libvpx-vp9), VP8 (libvpx), AV1 (libaom-av1), MPEG-4 (mpeg4) and copy. WebM offers VP9/VP8/AV1 and copy. H.265 and AV1 can compress more efficiently but need compatible players. copy keeps the picture without resolution, FPS or deinterlacing changes.
Audio codec
Initially aac, with Audio kbps at 128; range 16–512 kbps. MP4 offers aac/mp3, WebM opus/vorbis, MOV also pcm_s16le, AVI mp3/pcm_s16le; MKV also offers flac. copy retains audio, none removes it. Sample rate: 8000–192000 Hz; Channels: 1–8, both initially empty.
Width, Height and Scale
Width: 2–7680 px; Height: 2–4320 px. Odd values round down to even. Both fields start empty; one explicit side lets the other follow proportionally. Scale offers lanczos (default), bicubic, bilinear, fast_bilinear, neighbor and area.
FPS
1–240 fps; empty keeps the source rate. Fewer frames can save data but make fast movement less finely spaced.
Encoder preset
H.264/H.265 encoder speed, initially slow. Choices: ultrafast, superfast, veryfast, faster, fast, medium, slow, slower, veryslow. Slower choices spend more time finding compression at the same quality.
Tune, profile and level
Initially Auto. Tune: film, animation, grain, stillimage, fastdecode, zerolatency, psnr, ssim. Profile for H.264/H.265: baseline, main, high, high10, high422, high444, main10, mainstillpicture, matching the codec. Level: 3, 3.1, 3.2, 4, 4.1, 4.2, 5, 5.1, 5.2, 6, 6.1, 6.2. Keep Auto unless your target requires specific values.
Pixel format
Initially yuv420p; alternatives are Auto, yuv420p10le, yuv422p, yuv444p, yuv444p10le, yuv422p10le. A 10-bit format can retain finer gradients but needs a compatible encoder and player.
GOP and B-frames
GOP: 1–1000 frames between keyframes; shorter intervals help seeking. B-frames: 0–16, depending on codec. Both start empty; change them for specific output requirements.
Fast start
On by default for MP4 and MOV. It moves the index to the beginning of the file so playback can start before the download has finished.
Strip metadata
On by default. Removes tags such as recording device, creation time and location data that phones often embed.
Deinterlace, two-pass, remove audio
All initially off. Deinterlace cleans comb artifacts from interlaced recordings. Two-pass improves bitrate allocation in bitrate mode. Remove audio drops every soundtrack.

How much quality does a size limit leave you?

Target size mode works with simple arithmetic, and it helps to do it before you choose a resolution. The tool counts 1 MB as 1,048,576 bytes. A 10 MB target for a 60-second clip gives about 1,398 kbps in total; after 128 kbps of audio, roughly 1,270 kbps remain for the picture.

The same budget has to cover more detail at 1080p or 4K than at 720p, and fast movement needs more data too. With a small budget, lower the resolution in Advanced encoder and compare the result.

Leave a few percent of headroom below hard limits. The container adds a small overhead, and two-pass encoding lands close to the target but not on the exact byte.

Formats and limits

Input formats
MP4, MOV, MKV, WebM, AVI, M4V, MPEG/MPG, OGV
Output formats
Same as input, MP4, WebM, MKV, MOV or AVI
Maximum file size
500 MB per video; 1 GB on paid plans
Files per job
1
Processing
On our servers with FFmpeg, without an external service. Up to 15 minutes per job; demanding codecs such as AV1 on long 4K clips can reach that limit.
Watermark
None
Storage
Uploads and results are deleted automatically after 48 hours for guests and after 30 days for free accounts. Paid plans can keep files longer.

Frequently asked questions

Why is my compressed video larger than the original?

An efficiently compressed source can grow when encoded again. For example, changing H.265 to H.264 at a low CRF can require more bits. Choose Target size, raise CRF, keep H.265 or lower the resolution. Lossless output can also be large.

Does compression reduce quality?

Lossy encoding discards detail; Lossless avoids additional picture-compression loss. Resolution, pixel format and audio encoding remain separate choices. H.264 CRF 23 is the encoder default; 17–18 keeps more detail. Compare movement, dark scenes and fine textures.

Which settings suit common players?

MP4 with H.264 video, AAC audio, yuv420p and Fast start is a widely used combination. Check playback on your target device, especially if you choose H.265 or a 10-bit pixel format instead.

Can I compress without re-encoding?

Choose copy under Video codec with a container compatible with the source codec. Keep dimensions and FPS unchanged and Deinterlace off: the picture stream is copied without encoding. Audio can be copied, encoded or removed separately. Metadata, retained tracks and the container also affect file size; a smaller result is not guaranteed. To shorten the duration, use Cut Video.

What else makes a video smaller?

Shorter clips and lower resolutions have the biggest effect. Resize Video changes dimensions with more presets, and Mute Video removes the soundtrack without touching the picture.