View My GitHub Profile


Unum.DB is a high-performance persistent ACID database that supports a wide variety of workloads:

Why Unum?

  1. Fastest on the Market! 10x-100x performance improvements! Also more compact, than most DBs!
  2. No Vendor Lock! Easy to try & go back if you don’t like it! No new languages to learn!
  3. Extreme Flexibility! Runs on servers, phones or even IoT! No JVM required!
  4. Broad Functionality out of the box and easy integration with most common AI tools!

Fastest on the Market!

All benchmarks are pubulicly available. They are repeatable and include thousands of data-points. UnumDB outperforms all competitors in all workloads!

We usually beat the 2nd best result by 10x! That means 90% savings on your cloud infrastructure or 10x faster service for your customers. Here is how we reach those numbers.

No Vendor Lock!

Most DBs aren’t compatible with each other, and we don’t like that.
For every kind of workload (graphs, text, tables), we provide generic Python interfaces.
If you are not satisfied with our product, you can replace the backend implementation to something more familiar (like MongoDB, ElasticSearch or PostgreSQL).
Check out all the options here.

Extreme Flexibility!

Every aspect of UnumDB is adjustable:

Broad Functionality

The last point is particularly important! It guarantees that the functionality of UnumDB won’t limit you. You can always implement a custom pipeline using third party products and swiftly stitch it with UnumDB output.

How can it be so FAST?

Below are some of the bottlenecks we have identified in most modern DBs.
If you decide to write your own, those are the points to consider.





Interested? Get in touch for a demo!