Store and move a tenth of the data.
Compress uses learned codecs to shrink scientific, sensor, image and video data far beyond zip, with an error bound you set, or none at all.
- smaller than raw
- 10–50×
- smaller than raw
- or error-bounded
- Lossless
- or error-bounded
- decode speed
- GB/s
- decode speed
Features
What Compress does.
Neural codecs
Codecs trained on your kind of data.
Learn moreError bounds
Set the error per variable; we guarantee it.
Learn moreScientific formats
NetCDF, GRIB, Zarr, GeoTIFF and more.
Learn moreEdge & streaming
Compress at the sensor, before the link.
Learn moreSDK & CLI
One line in Python, a command in the shell.
Learn moreCloud, on-premises or air-gapped
Runs where your data already lives.
Learn moreHow it works
Three steps to smaller data.
01
Point it at your data
NetCDF, GRIB, Zarr, GeoTIFF, imagery, audio, video or streams, from the SDK, the command line or your pipeline.
02
Set the error you’ll accept
Lossless, or a bound per variable. Compress picks the codec and checks the bound on every file.
03
Store and move less
Files read like the originals, slice by slice, from cloud storage, a cluster or the edge.
Use cases
Who uses Compress.
Calculators
What would you save?
Your storage bill, your transfer times, your satellite passes: rough numbers in a minute.
Case study
Shrinking a national climate archive.
How a climate research centre cut its archive from 2.1 PB to 140 TB with error-bounded compression, and kept every result reproducible.
- smaller archive
- 15×
- smaller archive
- largest error on temperature
- 0.01 K
- largest error on temperature
- faster reads for analysis
- 3×
- faster reads for analysis
Latest articles
- News
Flowershift launches Compress
· 2 min read
- Scientific data
What an error bound actually guarantees
· 6 min read
- Codecs
Learned codecs against zstd, SZ and ZFP
· 9 min read
- News
GRIB2 and Zarr v3 support in the Compress SDK
· 2 min read
- Edge & space
Compressing on orbit: more scenes through the same pass
· 7 min read
- Storage & cost
The real cost of keeping a petabyte
· 5 min read
- News
Research partnership on compressing climate model output
· 2 min read
- Research
Will your files still decode in ten years?
· 6 min read
- Edge & space
Sensor data over thin links
· 4 min read
Questions.
What is Compress?
Is it lossy?
How much would we save?
Who makes it?
See it on your own data.
Every demo is run by the people who build Compress, on a sample of your own files.