How to Compress a Video for Free Online — MP4, MOV, AVI, WebM

Raw video is absurdly large. A single frame of uncompressed 1080p video is about 6 MB. At 30 frames per second, that's 180 MB per second, or roughly 10 GB per minute. Nobody actually works with raw uncompressed video outside of film production, but even the "compressed" files your phone produces are big because modern phones shoot at high bitrates to preserve detail.

An iPhone 15 Pro shooting 4K at 30fps in HEVC creates files around 40-50 MB per minute. Switch to ProRes and you're looking at 6 GB per minute. Samsung phones shooting in H.264 tend to produce even larger files than iPhones because H.264 isn't as efficient as HEVC at the same quality level.

The codec (H.264, H.265/HEVC, VP9, AV1) determines how aggressively the video can be compressed. The container (MP4, MOV, MKV, WebM) is just the wrapper — it holds the compressed video and audio streams together, but the container itself doesn't affect the file size in any meaningful way.

Video compression works by finding redundancy and eliminating it. There are two types of redundancy that codecs exploit:

Spatial redundancy — within a single frame, large areas of similar color (a blue sky, a white wall) can be described once rather than pixel by pixel. This is the same idea behind JPEG compression for images.

Temporal redundancy — between frames, most of the image stays the same. If you're filming a talking head, the background doesn't change at all. The codec stores the first frame fully (an I-frame), then subsequent frames only record what changed (P-frames and B-frames). A 30-second clip of someone talking at a desk might have 95% of its pixels identical across consecutive frames.

When you lower the bitrate, the codec has less data to work with, so it approximates more aggressively. Fine textures like hair, grass, and fabric are the first casualties. Fast motion gets blocky. Go too far and you get the classic "compression artifacts" — macroblocking, banding in gradients, and mosquito noise around edges.

Open the iFormat Video Compressor and drop your file in. You'll see a quality slider — that's your main control. Here's what the different levels actually mean in practice:

High quality (slight compression) — reduces file size by 30-50%. Visually indistinguishable from the original on anything smaller than a 27-inch monitor. Good for archiving footage you want to keep looking sharp.

Medium quality (balanced) — reduces file size by 60-80%. This is the sweet spot for sharing. Looks perfectly fine on phones and laptops. You'd need to pause and zoom in to spot differences from the original.

Lower quality (maximum compression) — reduces file size by 85-95%. Visible quality loss, especially in dark scenes and fast motion. Fine for previews, rough cuts, or getting under a tight file size limit.

Bitrate is the single biggest factor. It controls how much data per second the codec can use. YouTube recommends 8 Mbps for 1080p H.264 uploads and 12 Mbps for 4K. For sharing via email or messaging, you can go much lower — 2-4 Mbps for 1080p looks surprisingly good in H.264, and even better in H.265.

Codec choice matters a lot. H.265 (HEVC) produces files about 40-50% smaller than H.264 at the same visual quality. AV1 is even better but encoding is slower. H.264 has the widest compatibility — every device made in the last 15 years can play it. H.265 works on most modern devices but some older Android phones and pre-2017 computers struggle with it.

Frame rate is often overlooked. If your source is 60fps but the content is a presentation or talking head, dropping to 30fps halves the data with zero visual impact. You only need 60fps for fast motion — sports, gaming, action footage.

Audio takes up a tiny fraction of the file size (usually 128-256 kbps vs several Mbps for video), so compressing audio aggressively saves almost nothing. Leave it at 128 kbps AAC unless you need to hit an extremely tight size target.

These are the limits that send people looking for a compressor in the first place:

Email (Gmail, Outlook) — 25 MB attachment limit. A 1-minute 1080p clip from a phone is typically 100-150 MB, so you need significant compression or a shorter clip.

Discord — 25 MB without Nitro, 50 MB with Nitro, 500 MB with Nitro Basic. The free tier is brutal for video.

WhatsApp — 16 MB for video (bumped to 2 GB in some regions for document sharing, but the video player caps at 16 MB). WhatsApp also re-compresses your video on top of whatever you did.

Twitter/X — 512 MB for paid accounts, but the platform re-encodes everything anyway. Upload the highest quality version you can.

Upload portals — job applications, insurance claims, government forms. These vary wildly — anywhere from 5 MB to 100 MB. Check the limit before you start compressing.

For very large files (over 500 MB) or batch processing, desktop tools give you more control. HandBrake is free, open-source, and runs on Windows, Mac, and Linux. Its "Fast 1080p30" preset is a solid starting point — it uses H.264 with a CRF (Constant Rate Factor) of 22, which gives a good quality-to-size ratio.

If you want H.265 in HandBrake, switch the encoder to "H.265 (x265)" and bump the CRF to 24-26. The files will be smaller at equivalent quality, but encoding takes 2-3x longer.

FFmpeg is the command-line tool that powers most video processing software (including many online converters). It's incredibly powerful but the syntax is intimidating. A typical compression command looks like:
ffmpeg -i input.mp4 -c:v libx264 -crf 23 -preset medium -c:a aac -b:a 128k output.mp4

For quick, one-off compressions where you don't want to install anything, the online compressor is faster. For regular batch work on large files, HandBrake or FFmpeg is the better call.

Strip the audio if you don't need it. A screencast where you're going to add a voiceover later doesn't need the original mic audio. Removing the audio track can save 10-15% of the file size.

Trim before compressing. If your video has a 30-second intro of dead air or an accidental recording at the end, cut that out first. Less footage = smaller file = faster compression. You can use the iFormat Video Converter to trim and convert in one step.

Match the resolution to the delivery platform. Instagram Reels max out at 1080x1920. LinkedIn accepts up to 4K but most viewers are on phones. There's rarely a reason to upload 4K to any social platform.

Does compressing a video reduce quality?
Technically yes — compression always discards some data. But at reasonable settings, the difference is invisible in normal viewing. The goal is to find the point where the file is small enough and the quality loss is imperceptible.

Can I compress a video without losing quality at all?
Not really. You can re-encode with a more efficient codec (say, H.264 to H.265) which gives a smaller file at the same quality, but there's always a generation loss when re-encoding. The only truly lossless option is lossless codecs like FFV1 or HuffYUV, which produce files larger than most compressed formats.

What's the best format for compressed video?
MP4 with H.264 video and AAC audio. It plays everywhere. If you know your audience has modern devices, MP4 with H.265 gives you smaller files. See our MOV to MP4 guide if you need to convert first.

Why is my screen recording so large?
Screen recording apps often use low-compression codecs for performance — they need to encode in real time without dropping frames. macOS screen recordings use HEVC by default now, but OBS defaults to high-bitrate H.264. Re-encoding the finished recording at a lower bitrate shrinks it dramatically.

How small can I make a 1-minute video?
A 1-minute 1080p video can reasonably go as low as 5-10 MB in H.264 at 1-2 Mbps for talking-head content. For high-motion content like gaming footage, 15-20 MB is more realistic before quality gets noticeably bad.