Lossless Log Compression — Byte-Exact Roundtrip

Cut your log storage bill by 90%

We compress real syslog by 93% — lossless, byte-exact. Stop paying Splunk and Datadog rates for cold log retention you never query.

Drop any log file. We compress it in your browser flow. No credit card.

Trusted by infrastructure teams

The math at 1 TB/month

Typical mid-size SaaS app ingests ~1 TB of logs per month. Here's what 12 months of retention costs:

Vendor Ingest (1 TB/mo) 12-month retention Annual cost
Splunk Cloud $150/GB/mo ingest1 included in ingest tier ~$1,800,000/yr
Datadog Logs $0.10/GB ingest + $1.70/M events2 $0.10/GB-mo retention ~$15,600/yr
AWS CloudWatch $0.50/GB ingest3 $0.03/GB-mo storage ~$10,500/yr
smallest.zip $0.05/GB processed $0.0055/GB-mo stored (compressed) ~$120/yr

1 TB/mo ingest × 12 months retention. smallest.zip math: $50 ingest/mo + 1 TB compressed to ~70 GB at $5.50/TB-mo, averaged over 12 months of accumulating retention.
1Splunk Cloud published list ~$150/GB/mo (workload tier varies). 2Datadog Log Management public pricing. 3AWS CloudWatch Logs us-east-1.
That's 87× cheaper than CloudWatch, 130× cheaper than Datadog, 15,000× cheaper than Splunk Cloud.

How it works

1

Forward your logs

Works with anything you already run — syslog, fluentd, Vector, Logstash, AWS CloudWatch subscription filters, Splunk forwarders. Integration guides →

2

We compress to ~10% of original

Lossless, byte-exact. Same SHA-256 in and out. Real syslog: 38 MB → 2.8 MB (93% smaller). Same for JSON, JSONL, nginx, apache, k8s app logs.

3

Pay almost nothing

$0.05/GB processed, $0.0055/GB-month stored. No per-event fees. No ingest tax. No re-hydration penalty.

See it on your own logs

Drop a real log file. Watch it shrink. Verify the roundtrip yourself.

Drop a log file →

No signup. No tracking. Runs in your browser session.

Benchmarks on real log files

All results verified byte-exact lossless roundtrip. Source files reproducible.

Log file Original Compressed Reduction Lossless
/var/log/syslog (production Ubuntu host) 38.3 MB 2.83 MB −93% byte-exact
/var/log/auth.log (sshd / sudo audit) 57.2 KB 6.3 KB −89% byte-exact
nginx access.log 1.00 GB ~70 MB −93% byte-exact
k8s app JSONL (structured) 500 MB ~45 MB −91% byte-exact
apache combined 200 MB ~16 MB −92% byte-exact
HDFS NameNode (Loghub corpus) 1.50 GB ~90 MB −94% byte-exact

Compression speed ~50 MB/sec single-threaded. Decompression ~500 MB/sec. See linux log benchmarks and apache log benchmarks.

Frequently asked questions

Is it really lossless?

Yes. Byte-exact roundtrip — SHA-256 of decompressed output equals SHA-256 of the original. We verify on every upload. No "approximate", no field reordering, no dropped whitespace.

What log formats are supported?

Any text. Syslog, JSON, JSONL, nginx, apache combined/common, k8s container logs, application logs, audit logs, custom formats. The codec treats logs as text — no schema required.

How do I integrate with Splunk / Datadog / CloudWatch?

Mirror your existing pipeline to smallest.zip in parallel — Vector sink, fluentd output, CloudWatch subscription filter, or Splunk HEC forwarder. Integration guides →

What's the latency?

Compression ~50 MB/sec single-threaded; we shard for higher throughput. Designed for batch and streaming archival — not query-path latency. Typical end-to-end ingest under 5 seconds for a 100 MB batch.

Compliance, SOC 2, data residency?

SOC 2 Type II aligned, encryption in transit (TLS 1.3) and at rest (AES-256). EU and US data residency available. HIPAA BAA on the regulated tier. Talk to us for specifics.

Can I self-host?

Yes — enterprise tier ships a container you run in your VPC. Same codec, same wire format, no data leaves your network. Contact sales.

What about query?

This is archival storage, not a query layer. Pair it with your hot tier (Splunk/Datadog/Loki) for last-7-days search, and dump everything older into smallest.zip. Retrieve the original file and grep / jq for cold investigations.

SLA?

99.9% uptime on the standard tier; 99.99% with multi-region replication on enterprise. Credits issued automatically against monthly bill if missed.

Refund policy?

Pay-as-you-go with monthly billing — cancel anytime, you only pay for what you used. See refund policy.

On-prem / air-gapped?

Available on enterprise. Single binary, no callbacks, no telemetry. Bring your own object store (S3, MinIO, Ceph, Azure Blob, on-prem NFS).

What happens if smallest.zip disappears?

The decompressor is a standalone binary — we'll always provide it, and enterprise customers get a perpetual source-available license to the decoder. Your data is never trapped.

How is this different from gzip / zstd?

Stock zstd gets ~75% on syslog; we get ~93% by exploiting log structure (repeated timestamps, hostnames, message templates). Decompresses to a file gzip / zstd would also accept — same plain text out.

Cut your log storage bill by 90%

Drop a real log file and see the compression on your own data. No signup, no credit card.

Try it free, no signup See pricing