Video Compressor: Reduce Video File Size Free Online, No Upload

How CRF compression actually works

Unlike a fixed bitrate target, CRF adjusts the bitrate scene by scene to hold a consistent visual quality throughout the video. A static talking-head shot needs far fewer bits than a fast action sequence to look equally clean, so CRF spends less data where the picture is simple and more where motion or detail demands it. The result is a smaller average file size than a flat bitrate approach would produce at the same perceived quality, which is why CRF has become the default choice for general-purpose compression rather than manually picking a target bitrate.

How presets work

Each step up in CRF roughly doubles the perceptible quality loss while cutting the file size further, which is why the presets aren't evenly spaced. Balanced sits at the point most people settle on because it produces a file most viewers can't distinguish from the original at normal viewing sizes, while still shrinking a typical video by half or more.

Resolution as a second lever

CRF controls quality per pixel, but the pixel count itself is the other big factor in file size. Downscaling a 4K source to 1080p or 720p cuts the raw pixel count dramatically before compression even starts, which is why combining a resolution cap with a moderate CRF often beats pushing CRF higher alone. The resolution options here only scale down — a video already at 720p won't be stretched up to 1080p, since upscaling adds file size without adding real detail.

Single file vs. batch

The single-file tab is built for when you want to see the before-and-after size comparison and download one result. The batch tab is for clearing out a folder of recordings or exports at once — queue up several files, apply one set of preset and resolution settings, and let them process in sequence while you do something else. Each file in the batch reports its own progress and download link independently, so a slow file at the back of the queue doesn't block you from grabbing results that finished earlier.

Tips for best results

Start with Balanced and only move to a more aggressive preset if the resulting file is still too large — it's easier to compress further than to recover quality you already threw away. For screen recordings and tutorials, a resolution cap of 1080p combined with Small File usually looks fine, since text and UI elements compress differently than natural video and tolerate more aggressive settings. For footage you plan to edit further down the line, stick with High Quality or skip compression until after editing, since every re-encode introduces a small amount of generational quality loss.

Processing stays on your device

FFmpeg runs inside your browser through WebAssembly, so the video file is never uploaded anywhere. The trade-off is that speed depends entirely on your own CPU rather than a server farm built for this exact job — a large file on an older laptop will take noticeably longer than the same file on a recent desktop. In exchange, nothing about your footage passes through a third party at any point in the process.

Reading the size comparison

Once a file finishes, the result panel shows the original size, the compressed size, and the percentage saved between them side by side. That percentage is the fastest way to judge whether a preset was aggressive enough for your purpose — if a 500 MB screen recording only drops to 400 MB under Balanced, the source was likely already compressed efficiently, and jumping to Small File or Tiny will get you further than nudging the resolution down instead. If the savings look small across every preset, the original file's codec and bitrate were probably already close to what H.264 at a similar quality level would produce.

Is the Video Compressor free?

Are my video files uploaded to a server?

Do I need an account?

Which quality preset should I use?

Why is compression slow?

Why is the compressed file sometimes larger than the original?