Works with every major provider
Three steps from setup to automated billing.
Clients use the OpenAI SDK unchanged. You decide which provider handles each request — swap models without touching client code.
Every request is logged with token counts, latency, model, and computed cost. Full audit trail, always queryable.
Monthly invoices are generated with your markup applied and emailed to clients automatically. No manual reconciliation.
Built specifically for consultancies that bill AI usage to clients.
Drop-in replacement. Clients update one base URL and every OpenAI SDK call routes through Viapi — no code changes required.
Apply percentage markups, flat fees, or tiered rates per client. Invoices are generated and emailed automatically each month.
Watch revenue, token consumption, and provider costs update live. Drill down by client, model, or time range.
Set soft warnings and hard cutoffs per client. When a budget is hit, requests are blocked before you're on the hook for the cost.
Circuit breaker with automatic failover across OpenAI, Anthropic, Google, and xAI. Keep clients running when a provider has an outage.
Provider API keys are encrypted at rest with AES-256-GCM. Keys are never logged, never stored in plaintext, never exposed to clients.
Revenue, costs, and margins — per client, per provider, per day.
MRR
$12,480
+18%
Requests
847K
+31%
AI Cost
$3,210
+22%
Margin
74%
+4pp
Revenue vs. AI Cost — Last 30 Days
Top Clients by Revenue
Start free. Upgrade when you're billing clients.
Get started with zero upfront cost.