Your database.
Ready in 30 seconds.

CobraDB

Built different.

Six principles we don't compromise on.

Your bill follows your reality.

You pay for the gigabytes your data actually occupies on disk. Not what was allocated, not what you estimated, not a rounded-up tier. If your database shrinks, your invoice shrinks with it. Usage is measured hourly and aggregated monthly. No surprises, no minimums, no fine print.

It scales. You do nothing.

Most platforms make you choose a plan before you know what you need. CobraDB removes that decision. Your database grows from 1 MB to 100 TB on the same infrastructure, automatically. No migration windows, no support tickets, no plan changes, no downtime. It simply follows you wherever you go.

Queries arrive before your users blink.

Sub-millisecond latency is not a marketing claim. CobraDB deploys read replicas in the region closest to your users so every query takes the shortest possible path. When infrastructure is invisible, your product feels instant. We measure latency continuously. If it slips, we alert before you do.

Encrypted. Full stop.

Your data is protected by AES-256 at rest and TLS 1.3 in transit. Encryption is not a feature you enable. It is on by default, always, for every engine, every plan. What is stored on our infrastructure is unreadable.

Nothing is ever truly lost.

CobraDB takes an automatic snapshot every 60 seconds, not once a day like most providers. Each snapshot is replicated across multiple locations before we acknowledge it. Restore your database to any second in the past 30 days directly from your dashboard, without a support ticket, without waiting.

One URL. Your stack stays intact.

Every engine on CobraDB exposes fully compatible connection strings. Your existing ORM, your existing migrations, your existing drivers all continue to work without modification. Change your DATABASE_URL environment variable and your application is live on CobraDB. No rewrite, no risk, no learning curve.


Eight engines. One platform.

Choose your technology. We handle deployment, security and backups.

Relational

The gold standard for relational databases. Handles complex queries, foreign keys, full-text search and native JSON columns. Ideal for SaaS apps, financial systems and anything that demands data integrity.

postgresql.org →
Relational

The world's most deployed open-source database. Reliable, fast and universally supported. Powers WordPress, Drupal and most traditional web stacks.

mysql.com →
Document

Store documents exactly as your application shapes them. No migrations when your schema evolves. Perfect for product catalogs, user profiles and any data that doesn't fit neatly into rows and columns.

mongodb.com →
Columnar
Fastest

Columnar storage built for analytical queries. Ingests millions of events per second and returns aggregate results in milliseconds, even across billions of rows. The engine behind real-time dashboards and business intelligence at any scale.

clickhouse.com →
Search

Distributed search and analytics built on Apache Lucene. Index millions of documents, query them in milliseconds and visualize results with built-in dashboards. The backbone of search bars, log aggregation and security monitoring.

opensearch.org →
Time-series

PostgreSQL extended for time-series data. Automatically partitions by time, compresses historical records and accelerates time-range queries. Ideal for application metrics, sensor readings and financial tick data.

timescale.com →
Vector

The database built for AI. Store and query vector embeddings produced by LLMs, image models and audio encoders. Power semantic search, recommendation engines and retrieval-augmented generation with sub-millisecond similarity lookups.

trychroma.com →

We read every
message.

Because when something breaks, you need an answer. Not a knowledge base, not a chatbot, not a ticket that vanishes for a week. A person who knows the system and writes back.

See our plans →

2 days. Not 2 weeks.

Every request gets a real answer within 2 business days. Whether you are debugging a slow query or setting up a new engine, we are on it.

Engineers who built the thing.

The people answering you are the same people running the infrastructure. They know exactly what's happening because they built it.

Nothing is off-limits.

Schema design, slow queries, replication, migrations, billing. If it touches your database, we are happy to help. No topic is too small.

Your plan size doesn't change who answers.

We don't save the good engineers for big contracts. Every customer gets the same people, the same response time, the same care.


Pay for what you use.

Billed per GB actually stored. No pre-allocated infrastructure. No minimum commitment.

Free Get started.
€0.00 / GB / month (free)
  • 1 database
  • Up to 20 GB
  • 50 OPS*
  • Full encryption
  • Community support
Try for free →
Nano For small apps.
€0.009 / GB / month
  • Up to 100 databases
  • Up to 1 TB
  • 250 OP/s*
  • Daily snapshots
  • Full encryption
  • Support included
Choose Nano →
Enterprise
Custom dedicated infrastructure
  • Isolated infrastructure
  • 99.99% contractual SLA
  • 24/7 support
  • GDPR compliance
Contact us →

Everything you need to know.

How does per-storage billing work?

Every hour, we take a look at how much your database actually weighs on disk. We round up to the nearest gigabyte: 1.2 GB counts as 2 GB for that hour, and 0.8 GB counts as 1 GB. Simple. We then do that 720 times across the month and add it all up.

The result is your invoice. No pre-allocated storage, no block you pay for whether you use it or not. If you run a heavy migration on Monday and clean up half the data on Tuesday, Tuesday is cheaper. The meter follows your data, not the other way around.

Can I use CobraDB with my existing framework?

Completely. Think of CobraDB as your existing database stack, just hosted remotely instead of on your own machine. Same engines, same drivers, same queries. The only difference is that your data is encrypted, replicated and backed up automatically. Your code does not know the difference.

What happens when I approach my plan's storage limit?

We keep an eye on usage and we will reach out before you hit the ceiling. No surprises, no sudden cutoff. When it makes sense to move to a higher plan, one of our engineers handles the transition on our end. Your data stays exactly where it is, nothing gets moved or lost. From your side, it is just a price change.

Are my data automatically backed up?

Yes, every day without exception. We keep the last 3 daily snapshots so you can roll back quickly if something goes wrong. On top of that, we also keep a snapshot at 7 days, one at 15 days, and one at 30 days. That gives you four safety nets spread over a month, all accessible directly from your dashboard with no support ticket needed.

Where is my data hosted?

In ISO 27001-certified European data centers. We automatically replicate your data to the infrastructure closest to your clients, so queries stay fast without any configuration on your part. No data ever leaves the European Union.

Is there a contract period or cancellation fee?

None. You can delete your database at any time from your dashboard, with no notice period and no cancellation fee. You only pay for hours actually used.


Your next
database.

Stop spending time on infrastructure. Every minute configuring a server is a minute not spent on your product. CobraDB gives you a production-ready database in seconds, so you can get back to what matters.

Create my database →