Cheapest managed database hosting

Managed Postgres, MySQL, Redis, and document stores ranked by entry price. The important split is not the number but the model: provisioned instances bill whether or not you query them, serverless engines scale to zero.

List prices checked

All 11, ranked

#NameProviderEntry planNotes
1TursoTurso$8.25/mo500 databases, 9 GB of storage, and 1 billion row reads per month
2MongoDB AtlasMongoDB$9.00/moM0 shared cluster with 512 MB, always free
3Google Cloud SQLGoogle$9.37/moManaged MySQL, Postgres, and SQL Server on Google Cloud, billed per instance-hour with automatic storage growth that is easy to leave running unnoticed.
4Upstash RedisUpstash$10.00/mo500,000 commands per month
5AWS RDS for PostgreSQLAmazon Web Services$12.40/mo750 hours per month of a db.t-class instance for 12 months
6Azure Database for PostgreSQLMicrosoft$12.41/mo750 hours per month of a burstable instance for 12 months
7Scaleway Managed DatabaseScaleway$13.00/moEU-resident managed Postgres and MySQL with high-availability options, priced below the hyperscalers on both instance hours and storage.
8Neon Serverless PostgresNeon$19.00/mo0.5 GB of storage and 190 compute hours per month
9SupabaseSupabase$25.00/mo500 MB database and 5 GB of egress per month
10CockroachDB CloudCockroach Labs$25.00/mo$0 basic tier with monthly request-unit and storage allowances
11PlanetScalePlanetScale$39.00/moNone; the Hobby tier was retired in April 2024

Source: Turso (May 2026) · MongoDB (May 2026) · Google (May 2026) · Upstash (May 2026) · Amazon Web Services (May 2026) · Microsoft (May 2026). Published list prices only — committed-use discounts, regional variation, and per-request charges are not included. Not yet independently re-checked against the vendor’s page.

Questions

What is the cheapest managed database hosting in 2026?
Turso from Turso tops this ranking at $8.25/mo. The full table below shows all 11 entries with the same measure applied to each.
How is this ranking calculated?
Managed Postgres, MySQL, Redis, and document stores ranked by entry price. The important split is not the number but the model: provisioned instances bill whether or not you query them, serverless engines scale to zero. Every figure is the provider's own published list price, linked in the source column and stamped with the month it was checked.
Why is the cheapest option not always the right one?
Because a single axis never captures the whole bill. A provider with the lowest storage rate may have the highest egress rate; a model with the cheapest input tokens may be the most expensive for agentic work. The comparison pages model both sides against identical workloads, which is the check worth doing before switching.

Other rankings