Compare the trade-offs.
See the complete picture.
Seven codecs across the broad archive. Fresh holdout and media studies shown separately. Size, throughput and memory where measured.
Your bytes.
The real comparison.
Run Saar’s actual TypeScript fast engine against Brotli (qualities 5 and 11), gzip and DEFLATE on identical input. Results are computed on the server when you press Run.
32 KiB maximum · Not the native full profile · No bzip2, XZ, LZ4 or Zstandard in this live runner; see recorded studies for those codecs.Native fast profile, final candidate frozen before this workload set was acquired. Saar before/current frames are byte-identical.
Four acquired files from a five-file plan frozen before acquisition; Wikimedia returned HTTP 403 for the fifth. Engineer-selected and not independently curated. Process launch and pipe transfer are included. Peak memory and in-memory kernel speed were not measured. Results support this workload set only.
Compression ratio
Original size ÷ encoded size
Encoding speed
Original bytes processed per second
Decoding speed
Original bytes recovered per second
Compression ratio = total original bytes ÷ total encoded bytes (1× means no size change). Aggregate speed = total original bytes ÷ sum of per-file times, in decimal MB/s. Peak RSS = maximum recorded per-file child decode RSS. Recorded comparisons are not a live feed or an independent ranking. Compare within each set; different revisions and datasets are kept separate.
| Codec / setting | Output bytes | Compression ratio | Output / source | Encode MB/s | Decode MB/s | Peak decode RSS | Exact recovery |
|---|---|---|---|---|---|---|---|
| Saar fast · before | 12,828,323 | 1.693× | 59.06% | 73.61 | 218.89 | Not measured | Verified |
| Saar fast · current | 12,828,323 | 1.693× | 59.06% | 78.84 | 220.45 | Not measured | Verified |
| Brotli -q 1 | 11,364,125 | 1.911× | 52.32% | 153.03 | 196.8 | Not measured | Verified |
| Brotli -q 5 | 10,611,320 | 2.047× | 48.85% | 78.93 | 212.34 | Not measured | Verified |
| Brotli -q 11 | 10,006,714 | 2.171× | 46.07% | 0.82 | 170.86 | Not measured | Verified |
| Zstandard -1 | 11,347,067 | 1.914× | 52.24% | 201.3 | 311.87 | Not measured | Verified |
| Zstandard -3 | 11,004,921 | 1.974× | 50.67% | 158.97 | 324.41 | Not measured | Verified |
| LZ4 -1 | 12,913,972 | 1.682× | 59.45% | 193.55 | 286.1 | Not measured | Verified |
Test context & limitations
Environment. macOS-27.0.1-arm64-arm-64bit-Mach-O
Trials. One warmup, then seven rotated serial whole-command trials per codec and file. Charts sum the per-file medians.
Limitations. Four acquired files from a five-file plan frozen before acquisition; Wikimedia returned HTTP 403 for the fifth. Engineer-selected and not independently curated. Process launch and pipe transfer are included. Peak memory and in-memory kernel speed were not measured. Results support this workload set only.
Report SHA-256. 5d792330153858b1451fdf3f5663585987a953134c7daf84ee95d5d023d8e2a9
- Brotli
- brotli 1.2.0
- Zstandard
- Zstandard CLI 1.5.7
- LZ4
- LZ4 CLI 1.10.0
These are HEOSSI’s internal measurements, not an independent certification, standards approval or industry ranking. Different suites use different encoder revisions and must not be merged into one leaderboard.
Independent context. At the source.
External results use different hardware, data and methods. They are not directly comparable with Saar’s internal runs. Saar has no independently verified ranking on these sources.
On-demand GitHub repository metadata. This is not a live compression result or a Saar ranking.
Beyond a single file.
Every byte recovered.
Recorded 11 October 2026 (SGT). Generated test input, recovered exactly in two interoperability checks.
Historical stream verification, not a natural-data benchmark or a TB/PB capacity test. Encoded size was not recorded, so no compression ratio is claimed for this test.