CSV to Gzip Compressor
Compress CSV files to gzip in the browser. Shows original size, compressed size, and compression ratio.
Compress CSV data to the gzip container format defined by RFC 1952 directly in your browser. Paste rows or upload a file, run the DEFLATE pipeline locally, and download a .csv.gz that is byte-identical to what gzip on the command line would produce. The tool reports original size, compressed size, and the ratio so you can decide whether to keep the smaller copy.
CSV files are plain text, which means they compress extremely well. A 100 MB analytics export full of repeated column values often shrinks to 10 to 20 MB after gzip, because DEFLATE (RFC 1951) replaces repeated byte sequences with short back-references and then Huffman-codes whatever is left. The gzip container (RFC 1952) wraps that DEFLATE stream with a tiny header, a CRC32, and the original size, which is exactly what command-line gzip and HTTP Content-Encoding: gzip use.
This compressor accepts either pasted text or an uploaded .csv file. It encodes the text as UTF-8, runs it through the pako implementation of zlib, and hands back the resulting byte array as a downloadable .csv.gz. The header is preserved, so decompressing on any platform restores the original bytes exactly. The stats panel shows the input size, the output size, the percentage saved, and the ratio, which is useful when you need to know whether the saved bytes are worth the extra decode step on the receiving end.
Typical wins: data warehouse exports, sensor logs, financial tick data, and any CSV with repeated string values. Numeric-only CSVs with high entropy compress less, but rarely by under 30 percent.
- 1
Provide CSV
Paste rows into the editor or click Upload to read a local .csv file. The text stays in your browser memory.
- 2
Compress
Click Compress with gzip. The text is UTF-8 encoded and run through pako, the same DEFLATE algorithm zlib and gzip use.
- 3
Download
Review the size stats, then download a .csv.gz file. Standard tools like gunzip and zcat decode it without any special flags.
Shrink analytics exports
Compress a CSV dump from a BI tool before emailing it or uploading it to object storage where bandwidth is billed.
Save space in S3 or GCS
Athena, BigQuery, and Snowflake all read gzipped CSVs directly. Compressing first cuts storage cost and load time.
Ship logs as attachments
Shrink a server log exported as CSV to fit inside an issue tracker attachment limit without truncating rows.
Test decompression pipelines
Generate a real .csv.gz with known content to validate an ingestion job that expects gzip-encoded uploads.
Does the file leave my computer?
No. Compression runs entirely in your browser using the pako library. There is no upload, no server, and no analytics on the file contents. You can verify this by opening DevTools and watching the Network tab while you compress.
Is the output compatible with command-line gzip?
Yes. The bytes follow RFC 1952, the same gzip container the gzip and gunzip commands produce. You can decompress a downloaded file with gunzip file.csv.gz or zcat file.csv.gz on any Unix system, and Python's gzip module reads it without modification.
How much can I expect to save?
Plain CSV with text columns usually compresses to 10 to 25 percent of its original size. Numeric CSVs with float-heavy columns are closer to 40 to 60 percent. The stats card shows the exact ratio after each run.
Is there a file size limit?
There is no hard cap, but very large files (multi-gigabyte) can run out of browser memory because the whole buffer is held at once. For files over about 500 MB, command-line gzip is a better choice.
Can I change the compression level?
The tool uses the default zlib level (6), which is the same default the gzip command uses. It strikes the standard balance between speed and ratio. If you need maximum compression, run gzip -9 locally on the downloaded .csv before this step.