Pricing & Plans
How Parslinks plans, quotas, and usage-based credits work
Pricing & Plans
Parslinks billing combines a plan (what you're entitled to) with usage-based credits (what you consume). Exact prices and limits are shown live in the dashboard — this page explains the model so you can predict your bill.
The model
- Your workspace is on exactly one plan at a time.
- Plans set entitlements: how many projects/services you can create, replica limits, build concurrency, storage caps.
- Running resources generate usage events (replica-hours, builds, stored GB, traffic) that consume workspace credits.
- Each plan includes a monthly credit allowance; you can top up any time.
What consumes credits
| Resource | Metered as |
|---|---|
| Running replicas | Replica-hours by region |
| Builds | Per build execution |
| Artifacts + volumes | Stored GB-month |
| Public bandwidth | GB served |
Idle things cost little: stopped instances don't burn replica-hours, and artifacts of old deployments are garbage-collected once nothing retains them.
Free tier
New workspaces start on the free plan: enough entitlement for a personal project with a production environment and previews, plus a starter credit allowance. Quota errors surface clearly when you hit the ceiling — upgrade or free up resources to continue.
Upgrading & downgrading
- Upgrade: new entitlements apply immediately; running workloads are unaffected except that new actions can now exceed the old quota.
- Downgrade: running resources keep running, but anything beyond the new limits blocks new creations/scales until you're back within quota.
- Plan changes prorate per the terms shown at checkout.
Checking your usage
The billing page shows live aggregates per project and service, plus the credit ledger. Because metering is event-based, every number on that page traces to recorded events — see Billing & Usage.
Tips to keep costs predictable
- Pause preview/custom environments you're not actively using
- Right-size placement intent — replicas you don't need still bill while ready
- Let artifact retention clean up superseded deployments instead of hoarding
- Set up low-credit awareness among admins (notifications cover it)