DocsResourcesRate Limits
Rate Limits & Quota
Understand request rate limiters, token-bucket ceilings, and credit balance consumption.
Token-Bucket Rate Limiting
To guarantee system reliability, VeriLens enforces rate limits across three independent boundaries:
Rate Limit Boundaries
Active rate limits across the platform:
| Endpoint Category | Rate Ceiling | Burst Capacity | Violation Penalty |
|---|---|---|---|
| Prediction Endpoints | 1 request / second | 5 requests | HTTP 429 Too Many Requests |
| Authentication Routes | 5 requests / minute per IP | 5 requests | 15-minute lock after 5 consecutive failures |
| General API Endpoints | 100 requests / minute | 20 requests | HTTP 429 Too Many Requests |
Credit Balance & Tier Pricing
• **Fast Check (`/api/v1/predict`)**: 0 credits. Free for all authenticated users. • **Deep Analysis (`/api/v1/predict/deep`)**: 1 credit per execution. Balance is shared across all API keys belonging to the account. • **Auto-Refund Policy**: If an error occurs during evidence retrieval or verdict generation, the credit is immediately returned to your account balance.