Measured - byte-exact - 2026

Pedulli vs 7-Zip

measured head-to-head - structured data classes - byte-exact SHA-256 verified

TL;DR - honest
Pedulli is a best-of-N racer: it races xz, zstd, brotli and your data's SRD math and keeps the smallest verified output, always byte-exact (SHA-256 roundtrip) with a +1 byte never-worse floor. So it is never larger than the best standard codec (worst case +1 byte): it wins outright on structured classes (JSON, periodic and sparse binary), and on already-optimal or random data it ties the best codec - on incompressible data that means +1 byte vs 7-Zip's hundreds. 7-Zip ships a polished desktop archive format and container - we say so plainly.

The measured table

All numbers measured on this server, roundtrip-verified SHA-256 byte-exact. Proofs available on request.

Input	7-Zip mx=9 (ultra)	Pedulli (best-of-N)	Δ
MP4 master 10 MB	9,468,640 B	9,430,108 B	-38,532 B
1 MiB of zeros	168 B	13 B	much smaller (redundant input)
1 MiB periodic binary	1,242 B	436 B	-65%
JSON 31 KB	4,201 B	1,265 B	-70%
HTML 161 KB	36,108 B	36,108 B	tie (selects best codec)
Apache logs 3.5 MB	213 KB	213 KB	tie (selects best codec)
Random bytes 1 MB	1,049,234 B (+658 B)	1,048,577 B (+1 B)	657 B less overhead

Where 7-Zip fits as a desktop tool (honest)

Long English text. 7-Zip's large LZMA dictionary is strong on plain prose over ~100 KB - so Pedulli races it as one of its internal coders and selects its output, tying 7-Zip on that class rather than ever coming out larger.
GUI + archive format. 7-Zip ships a polished Windows GUI and is both a codec and a multi-file archive container. Pedulli is a codec exposed via web + API.
Universal availability. 7-Zip is installed everywhere on the desktop.

What Pedulli does that 7-Zip does not

+1 byte never-worse floor. Pedulli's own output is never more than 1 byte larger than the input. On random data 7-Zip adds hundreds of bytes per file - that compounds at scale.
Best-of-N racing. Several internal coders run and the smallest byte-exact result is kept, so structured classes (JSON, periodic, sparse) come out tighter.
Web-native + API. In-browser for small inputs, REST API for pipelines (CI/CD, backups, log shipping). 7-Zip is CLI / GUI only.
Zero data retention. GDPR by design, EU-sovereign.

When to switch

Switch to Pedulli for video masters, JSON/XML telemetry archives, programmatic pipeline compression, and the per-file +1 byte never-worse floor - and because the best-of-N racer is never larger than the best standard codec on any class, including long English text (where it races and ties 7-Zip's dictionary). Keep 7-Zip when you want a desktop GUI and a multi-file archive container.

Run the free trial Pricing / API All benchmarks

More comparisons: gzip - zstd - xz - brotli - zpaq