Pricing
Gauntlet is pre-beta and nothing is for sale. This page describes the pricing model so you can judge whether it suits your workload. It is indicative and may change before launch.
Indicative, pre-launch. No rates, free allowances or commitments are published. Any number you see elsewhere is not a Gauntlet price.
No charge from us
Run the binary on your hosts against your bucket.
Rates not published
A managed service, designed and not built.
Rates not published
Gauntlet operated in your cloud account.
The managed model
Three line items, each tied to something you can see. There is no charge per request, per row read or per query.
| Line item | What it measures | Status |
|---|---|---|
| Plan | A flat amount per organisation for the service level, with a published commit-rate ceiling per plan and node size. | Rate not set |
| Allocated nodes | Node time, by node size. The compute you reserve, whether or not databases are active on it. | Rate not set |
| Live logical storage | The data your databases currently hold, measured daily. Not physical bytes, and not history you did not ask to keep. | Not billed until reclamation is on |
| Retained history | Point-in-time windows and branches, as a separate line once those features exist. | Coming |
An idle database makes no bucket requests, so its incremental cost is its stored bytes. It does not make a plan or an allocated node free. Many idle databases on one small node is the case this model is built for.
Each plan and node size will publish a commit-rate ceiling, because every commit is a write to the bucket. You will see the ceiling before you buy, not after you hit it.
Usage shown in a console is an estimate until the period closes. Finalised usage and estimated usage will be labelled separately. A bill is never computed from Query Insights counters.
The managed service's general-availability gate is that every claim on this page has a published measurement behind it. Until then this page describes intent.