Quotas, limits & pricing

Advania ALT enforces per-model, per-user rate limits at the API gateway and per-account TTLs on API keys. Pricing is per-model.

In this section

• Rate limits — how limits are applied and how to read them from responses
• API key TTLs — how long API keys stay valid
• Model pricing — per-token pricing by model

Endpoint-specific limits are exposed on responses via x-ratelimit-* headers — see the interactive OpenAPI for the current shape.