PlanetScale Pricing Calculator
Estimate your real monthly PlanetScale bill for your Postgres cluster — single-node or HA — and compare against other managed database providers.
1. Pick a workload
Choose the preset closest to your current or projected database usage.
2. Customize (optional)
Adjust storage, compute hours, and workload options.
PlanetScale Postgres
Estimated monthly cost · 100 GB storage · 2 GB RAM
Same workload, all databases
Estimated monthly cost — sorted cheapest first
Convex uses per-seat pricing — costs grow with team size. Think something is wrong?
What PlanetScale excels at
- Millisecond-level billing: clusters are billed to the millisecond — no wasted spending on idle resources.
- HA clusters: 3-node HA available from $15/month — zero single-point-of-failure.
- Dedicated PgBouncer: connection pooling as a separate managed instance, not shared.
- Multiple cloud regions: 24+ regions across AWS and GCP.
- No vendor lock-in: standard Postgres — migrate in and out with pg_dump/pg_restore.
How PlanetScale Postgres pricing works in 2026
PlanetScale Postgres uses cluster-based pricing: you pay a flat monthly fee for a dedicated cluster, billed to the millisecond and prorated daily. Clusters are available as single-node or 3-node HA configurations. Entry price is $5/month for a PS-5 single-node cluster (512 MB RAM, arm64). HA starts at $15/month for a PS-5 3-node cluster.
Storage is 10 GB included per cluster, then $0.125/GB/month overage. Egress is 100 GB/month included on production branches, then $0.06/GB. Pricing shown uses arm64 instances, which are 18–20% cheaper than equivalent x86-64 configurations.
PlanetScale cluster tiers
PS-5 (512 MB RAM, $5/$15 HA), PS-10 (1 GB RAM, $10/$30 HA), PS-80 (8 GB RAM, $49/$148 HA), PS-160 (16 GB RAM, $95/$286 HA), PS-320 (32 GB RAM, $190/$570 HA). There is no PS-2 or PS-4 tier — you jump from 1 GB directly to 8 GB. If your workload needs 2–4 GB RAM, you either use PS-10 with swap or upgrade to PS-80.
PlanetScale pricing examples
Startup SaaS (PS-10, single node): $10/month cluster + minimal storage and egress = $10/month.
Production SaaS (PS-80, HA): $148/month cluster + 5 GB storage overage × $0.125 + 50 GB egress within included = $148.63/month.
Scale (PS-160, HA, high egress): $286/month cluster + $12.50 storage overage + $6 egress overage = $304.50/month.
What makes PlanetScale different
PlanetScale's key strengths are millisecond billing, HA from $15/month, and a dedicated PgBouncer connection pooler available as an optional add-on. The cluster model gives you predictable costs without the variable compute billing of serverless databases — you know exactly what you're paying before the month starts.
PlanetScale operates 24+ regions across AWS and GCP, and uses standard Postgres — no vendor lock-in. You can migrate in with pg_dump and out the same way.
How this calculator works
The calculator selects the nearest PS cluster by RAM target. If HA is enabled, the 3-node cluster price is used. Storage overage (above 10 GB) and egress overage (above 100 GB) are added separately. The optional dedicated PgBouncer add-on ($18–$933/month) is not included.
Frequently asked questions
How much does PlanetScale Postgres cost?
Entry price is $5/month for a single-node PS-5 cluster. A typical production SaaS on PS-80 with HA runs $148/month. There is no free tier — PlanetScale targets production workloads from the start.
Does PlanetScale support high availability?
Yes. Every cluster size is available in a 3-node HA configuration. Failover is automatic. HA clusters cost approximately 3× the single-node equivalent — PS-10 HA is $30/month vs $10 single-node.
PlanetScale vs Neon — which is cheaper?
For always-on workloads with low RAM requirements, PlanetScale's flat cluster pricing (PS-10 at $10/month) often beats Neon's hourly compute model. For variable-load workloads with scale-to-zero, Neon is cheaper. The break-even point depends on your active hours per month.
Is PlanetScale standard Postgres?
Yes. PlanetScale Postgres is a managed Postgres offering — you connect with a standard Postgres connection string and use any ORM or driver. It replaced PlanetScale's original MySQL-based offering.
MakerKit is compatible with PlanetScale Postgres
Standard Postgres patterns work with PlanetScale. Connect your cluster and skip the boilerplate.
Compare with other databases: