Measured - byte-exact - 2026

Pedulli vs zstd

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 on its own output - so it is never larger than the best standard codec (worst case +1 byte). It wins outright on structured data (JSON, HTML, source, periodic, MP4) and ties the best codec on already-optimal/random data, where it adds +1 byte against zstd's 13-39. zstd is far faster - that is its design priority and it wins throughput. Different jobs.

The measured table

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

Input	zstd-19 / --long	Pedulli (best-of-N)	Δ
1 MiB of zeros	43 B	13 B	much smaller (redundant input)
1 MiB periodic binary	1,872 B	436 B	-77%
Apache logs 3.5 MB	240 KB	217 KB	-9.5%
JSON 31 KB	2,801 B	1,265 B	-54.8%
HTML 161 KB	41,288 B	38,757 B	-6.1%
MP4 master 10 MB	9,481,772 B	9,430,108 B	-51 KB
Random bytes 1 MB	1,048,589 B (+13 B)	1,048,577 B (+1 B)	12 B less overhead

What zstd does better (honest)

Pure throughput. zstd was designed for fast single-core streaming. For raw binary throughput zstd wins - keep it there.
Kernel integration. btrfs, ZFS and Linux compressed-RAM all use zstd. Pedulli is a service / library, not a kernel codec.
Dictionary training. zstd's `--train` mode adapts to your corpus offline.

What Pedulli does that zstd does not

+1 byte never-worse floor. Pedulli's own output is never more than 1 byte over the input. zstd adds 13-39 bytes on incompressible data.
Best-of-N racing. xz, zstd, brotli and your data's SRD math all run; the smallest byte-exact result wins, so Pedulli is never larger than the best standard codec (worst case +1 byte) and structured classes come out tighter.
Web-native + API. In-browser for small inputs, REST API for pipelines. zstd is a CLI / library.
Zero data retention. GDPR by design, EU-sovereign.

When to use which

Use zstd when throughput dominates (kernel-level, btrfs, ZFS, RPC compression) and ratio is secondary. Use Pedulli when ratio matters: log archives, JSON/XML telemetry, HTML/JS bundles, video masters, backups - plus the per-file +1 byte never-worse floor.

Run the free trial Pricing / API All benchmarks

More comparisons: gzip - xz - 7-Zip - brotli - lz4