About File Size Analyzer
File Size Analyzer takes a folder or selection of files and gives you a visual breakdown of where the bytes are going — totals by file extension, the biggest individual files, and a percentage split of disk use by type. Useful for cleaning up a downloads folder, auditing what's eating space in a project, or figuring out why an export is suddenly enormous. Everything is read locally; the files themselves are never transferred.
- No uploads
- Browser-only
- Works offline
- 100% free
How it works
- 1
Drop a folder or files
Pick a whole folder for a full directory analysis, or select individual files.
- 2
Watch the breakdown render
A pie chart and table appear, grouping bytes by extension (e.g. .mp4, .pdf, .png) and listing the top largest individual files.
- 3
Drill into the offenders
Sort by size to find the biggest space-hogs. Filter by extension to see which media types are consuming the most disk.
KB vs KiB: the 1000-versus-1024 confusion
There are two different 'kilobytes' in the wild and they don't match. The decimal definition (SI) says 1 KB = 1000 bytes, 1 MB = 1,000,000 bytes — this is what storage manufacturers print on the box and what macOS reports. The binary definition says a 'kibibyte' (KiB) = 1024 bytes, and a 'mebibyte' (MiB) = 1024 KiB — this is what Windows shows while still labelling it 'KB', and what RAM is always measured in. The IEC introduced KiB/MiB/GiB specifically to end this ambiguity (see the byte units reference).
The practical fallout: a drive sold as '1 TB' (1,000,000,000,000 bytes, decimal) shows up in Windows as about 931 'GB' (because Windows divides by 1024 cubed but writes GB). Nothing is missing and nobody cheated — the two systems just disagree by about 7% at the terabyte scale, and the gap widens the larger the unit. When a number looks 'short', this mismatch is the usual culprit.
| Unit | Decimal (SI) | Binary (IEC) | Where you see it |
|---|---|---|---|
| 1 KB / KiB | 1,000 bytes | 1,024 bytes | Decimal: macOS, drive labels |
| 1 MB / MiB | 1,000,000 bytes | 1,048,576 bytes | Binary: Windows, RAM |
| 1 GB / GiB | 1,000,000,000 bytes | 1,073,741,824 bytes | Gap is about 7.4% at this scale |
| 1 TB / TiB | one trillion bytes | 1,099,511,627,776 bytes | "1 TB" drive shows ~931 GiB |
Same file, two correct numbers — the only error is comparing a decimal figure against a binary one.
What actually inflates a file
- Images and media dominate. An uncompressed photo or a PNG of a photograph can be 10x a well-chosen JPEG; embedded video and audio dwarf text. A 'huge' document is almost always its pictures.
- Uncompressed or lossless formats trade size for fidelity: BMP, TIFF, WAV and RAW are large by design; PNG is lossless and bloats on photographic content where JPEG/WebP would be a fraction of the size.
- Embedded fonts, thumbnails and metadata add up — Office files and PDFs carry preview images and revision history you can't see.
- Already-compressed data doesn't shrink again: re-zipping JPEGs, MP3s or MP4s saves almost nothing because the entropy is already squeezed out.
Where the bytes hide (and what this won't tell you)
Two surprises trip people up. First, many file formats — DOCX, XLSX, PPTX, EPUB, JAR, even APK — are secretly ZIP archives. The bytes inside are already compressed, which is why their on-disk size barely changes if you zip them again, and why you can open them with an Extract ZIP tool to see what's eating the space. Second, 'size on disk' differs from 'size': filesystems allocate space in fixed blocks (often 4 KB), so a 100-byte file still consumes a full block — a folder of thousands of tiny files wastes far more disk than the sum of their byte counts suggests.
This analyser reports logical byte counts of the files you give it; it doesn't crawl folders for you (browsers can't read directories silently) or account for filesystem block overhead and compression on the drive itself. For trimming an oversized image specifically, pair it with an image compressor; to spot redundant copies eating space, a duplicate finder is the targeted tool.
Frequently asked questions about File Size Analyzer
What information does it read from my files?
Just the file name, file size, and last-modified timestamp — the same metadata your operating system shows in a directory listing. The actual file contents are never read, decoded, or processed. Browser file APIs intentionally expose this metadata without requiring a full file read.
Can it analyse a whole disk or just a folder?
Browsers can't access an entire disk for security reasons — they're sandboxed to folders you explicitly pick. For a whole-disk audit, use a desktop tool like WinDirStat (Windows), DaisyDisk (Mac) or ncdu (Linux). For folder-level audits this tool is perfect.
Does it follow symlinks or include hidden files?
Browser file pickers traverse normal directory entries only — symlinks are reported as their pointer, not followed recursively. Hidden files (those starting with a dot on macOS/Linux) are typically excluded by the OS-level file picker.
Privacy, offline use, browser support, and pricing questions are answered on the site-wide FAQ.