Reduce MP4 video file size while maintaining quality. Compress videos for social media, email, and storage.
Upload your MP4 file to compress
MP4 video compression reduces file size by applying efficient codecs (H.264, H.265, VP9) that remove redundant visual information. Our tool uses FFmpeg to re-encode videos with optimal settings for your target platform.
Video compression works through inter-frame prediction (storing only changes between frames), intra-frame compression (similar to JPEG within each frame), and entropy coding. The CRF (Constant Rate Factor) controls quality — lower values (18-23) preserve more detail, higher values (28-40) produce smaller files.
For social media, use H.264 at 1080p with medium quality. For archival, use H.265 at original resolution with high quality. For web delivery, VP9 offers excellent compression at the cost of slower encoding.
H.264 offers the broadest compatibility across all devices. H.265/HEVC provides ~50% better compression but requires newer hardware. VP9 is best for web playback on Chrome, Firefox, and Edge.
Typical compression ratios range from 40-80% depending on settings. Combining resolution downscaling (e.g., 4K to 1080p) with efficient codec settings can achieve 80-90% reduction.
CRF (Constant Rate Factor) controls quality. Range is 0-51. Low = high quality. For H.264: 18-20 (visually lossless), 23 (good quality), 28 (acceptable), 35+ (small file, low quality).
Our tool preserves the original audio track by default. You can optionally re-encode audio with AAC at various bitrates to further reduce file size.
Yes, use our dedicated social media presets at /social for platform-specific settings optimized for Instagram, YouTube, TikTok, Twitter, and more.
For archival, use H.265 at the original resolution with CRF 18-20. This preserves near-original quality while achieving 40-60% compression compared to the source.
MP4 video compression achieves file size reduction through a sophisticated combination of spatial and temporal compression techniques. The MP4 container format holds video compressed with codecs like H.264 (AVC), H.265 (HEVC), or VP9, alongside an AAC audio track. Temporal compression (inter-frame prediction) stores complete keyframes (I-frames) periodically, while storing only the differences between frames as P-frames (predicted) and B-frames (bidirectional). This is why videos with minimal motion compress much more efficiently than action sequences — each P-frame stores only the pixels that changed. Spatial compression (intra-frame) applies techniques similar to JPEG within each frame, using block-based DCT transforms and quantization. The key quality control is the CRF (Constant Rate Factor), a scale from 0-51 where lower values mean higher quality. CRF 18-20 is considered visually lossless, 23 is the default for good quality, 28 is acceptable for web delivery, and 35+ produces very small files with visible artifacts. Unlike bitrate-based encoding which uses a fixed data budget, CRF adapts the bitrate to the content complexity — allocating more data to complex scenes and less to static ones.
H.264 (AVC) is the most universally supported video codec — it plays on virtually every device and platform. It offers good compression with fast encoding speeds, making it ideal for web delivery, social media, and compatibility. H.265 (HEVC) delivers approximately 50% better compression than H.264 at the same quality, but requires hardware decoding support (iPhone 6+, Android 5+, PS5, Xbox Series X). It excels for 4K content, archival, and storage-constrained scenarios. VP9, developed by Google, offers compression efficiency between H.264 and H.265 with royalty-free licensing. It is the codec behind WebM and YouTube's recommended format, with excellent web browser support (Chrome, Firefox, Edge). VP9 encoding is significantly slower than H.264 but faster than H.265, making it a pragmatic middle-ground for web delivery.
Downscaling resolution is the single most effective way to reduce video file size. A 4K video (3840x2160) has four times the pixels of 1080p — downscaling to 1080p alone can reduce file size by 60-70% before codec compression is even applied. Combined with efficient CRF settings, total reduction of 80-90% from the original is achievable. Our tool offers resolution presets for 4K, 1080p, and 720p, along with four quality levels that map to specific CRF values: Very High (CRF 18), High (CRF 23), Medium (CRF 28), and Low (CRF 40).
| Codec | Compression vs H.264 | Encoding Speed | Device Support | Best For |
|---|---|---|---|---|
| H.264 (AVC) | Baseline | Fastest | 100% | Web, social media, universal playback |
| H.265 (HEVC) | ~50% better | Slower | Modern devices | 4K video, archival, storage savings |
| VP9 | ~30% better | Medium | Web browsers | Web delivery, YouTube, Chrome/Firefox |
| CRF Value | Quality Level | Typical Size Reduction | Use Case |
|---|---|---|---|
| 18-20 | Visually lossless | 20-40% | Archival, professional editing master |
| 21-23 | High quality | 40-60% | YouTube upload, video distribution |
| 24-28 | Good quality | 60-75% | Web streaming, social media |
| 29-35 | Acceptable | 75-85% | Mobile streaming, previews |
| 36-40 | Low quality | 85-90% | Maximum compression, small previews |
Always downscale to the display resolution your audience will use. A 4K source downscaled to 1080p for YouTube or social media saves massive bandwidth with no perceived quality loss for viewers on 1080p screens.
Two-pass encoding analyzes the video in a first pass to identify complex scenes, then distributes the bitrate budget optimally. This produces better quality at the same file size compared to single-pass encoding.
Audio tracks are often a significant portion of total video size. Re-encoding stereo AAC audio at 128kbps saves space with minimal quality loss. For voice-only content, 64kbps is often sufficient.
Keyframe interval (GOP size) affects seeking performance and compression. A 2-second interval (48-50 frames at 24fps) offers good seeking with efficient compression. Longer intervals improve compression but make seeking slower.
CRF (Constant Rate Factor) adaptively allocates bitrate based on scene complexity — simple scenes use fewer bits, complex scenes use more. Bitrate encoding uses a fixed data budget regardless of content, which can waste bits on simple scenes or under-allocate on complex ones. CRF is generally preferred for quality-focused compression.
For most viewers on typical screens, CRF 23 is indistinguishable from the source. It is the x264 default for good reason — it offers an excellent balance of file size and visual quality. Critical viewing or editing masters should use CRF 18-20.
Yes — our tool lets you downscale to 720p, 1080p, or keep the original 4K resolution. Downscaling is the most effective way to reduce file size, often achieving 60-70% reduction before codec compression.
For archival, use H.265 (HEVC) at the original resolution with CRF 18-20. This preserves near-original quality while achieving 40-60% compression compared to the source. Always keep an uncompressed or lightly compressed master copy.
Longer videos tend to compress more efficiently because the ratio of keyframes (which are expensive) to predicted frames improves. However, the absolute file size savings increase with length — compressing a 10-minute video saves more total space than compressing a 1-minute clip.