Rate Limits & Quotas
Pragence applies a deployment-level request-per-minute limit and an organization entitlement for daily API-key calls. Other quota values, including storage, documents, seats, and monthly conversations, come from the organization's resolved entitlements and can vary by deployment and offer. There is no repository-backed universal plan-limit table or fixed overage price.
Request limits
The FastAPI application configures a per-minute request limit from
RATE_LIMIT_REQUESTS_PER_MINUTE (default: 60). API-key authenticated requests
also check the organization's max_api_calls_per_day entitlement. When the
daily cap is reached, that API-key request returns 429 Too Many Requests and
includes a Retry-After header. The daily counter is per organization and uses
Redis when caching is enabled; if the counter is unavailable, the current
implementation fails open.
Usage and entitlements
Organization entitlements are returned by the billing API's organization
entitlements endpoint when billing is enabled. The usage endpoint reports
current-period document, user, query, and storage metrics. Consult the live
deployment's /openapi.json and entitlement response for exact routes and
limits. Billing and entitlement routes are omitted in self-hosted mode.
The repository does not define universal 50%, 80%, or 100% warning thresholds, conversation auto-recharge behavior, or a per-conversation overage price. Confirm commercial limits and any configured notifications with your administrator or provider.