- Who it helps
- Developers adding customer-level budgets to AI apps.
- Availability
- Early MIT-licensed 0.1.0 TypeScript library; README requires Node 18+. The optional hosted service has separate access terms.
Where the cap applies
The wrapper counts returned token usage using a price table. It can switch configured models near a monthly cap and reject later calls. The default in-memory counter is process-local and disappears on restart.
Concurrent calls and outages
A reserving store holds estimated cost before a call and settles actual usage afterward. The README still allows roughly one-call overshoot. Store failures allow requests by default; failOpen: false changes that policy. This is not an absolute invoice ceiling.
What you can do
- Read the linked primary evidence and its access limits before planning a trial. Open reference ↗
Where the evidence stops
README checked November 6. No installation, billing reconciliation or concurrent-request test executed. Provider pricing and hosted terms require separate checks.
Sources read. Sources read November 6, 2026 UTC. No third-party project, model inference or complete product workflow was executed.
Sources & further reading
GitHub · official
LLMBrake budgets need an explicit outage policy ↗
Primary page read November 6. Claims were checked against the page; the workflow was not executed.
LLMBrake · page checked 2026-11-06
Coverage note: Original YouTube demonstrations and independent cross-platform attention were not established. No end-to-end workflow was executed.
The record over time (1)
Dated source checks and published explanations. An observation date can differ from the original release date.
- · Added to Current
LLMBrake meters AI spending per customer, but its default outage behavior allows calls through.