Provider pricing

Requesty AI API pricing

Requesty is a hosted LLM gateway sitting in front of 250+ models from 30+ providers, adding semantic caching, automatic fallback, prompt-level cost analytics, and per-tenant rate/spend limits. Charges a small per-token markup on top of upstream provider rates plus a free observability tier.

Founded2024
HeadquartersNew York, NY
Tracked modelsDirectory

Best for

B2B SaaS shipping LLM features to many tenants who need spend control and observability.

Strengths

  • Smart routing across 250+ models with automatic fallback
  • Semantic caching cuts repeat-prompt cost by 30–60%
  • Per-tenant budgets and PII redaction for SaaS apps
  • Full prompt/response logging with no vendor lock-in