One endpoint. Every model.
Pay for what you use.
The commercial AI gateway: OpenAI-compatible API, prepaid wallet, project budgets, and adaptive routing across upstream channels.
curl $RELAY_BASE_URL/v1/chat/completions \
-H "Authorization: Bearer sk-rl-…" \
-d '{"model":"omnigent-auto","messages":[{"role":"user","content":"hello"}]}'OpenAI-compatible API
Point your existing SDK at one base URL. Chat completions, streaming SSE, and a curated public model catalog.
Prepaid wallet & budgets
Top up, watch every request settle against your balance in real time, and cap spend per project.
Adaptive channel routing
Requests route across official, compatible, BYOK and self-hosted channels with per-attempt failover.
Usage you can audit
Per-request metering facts — tokens, charge, latency, routing — never your prompt bodies.
Pricing
Sales prices per 1M tokens are loaded from the Relay catalog. We do not estimate or substitute prices when the catalog is unavailable.
Loading live catalog pricing…