Feature — Artificial Intelligence

AI Traffic Control Without Cloud Lock-In

Proxy OpenAI, Anthropic, or any local model through apiRuler. Apply prompt protection, token quotas, and cost limits — without changing a line of application code.

AI API proxy and management

Your AI APIs. Your Policies.

AI calls are expensive and hard to govern. apiRuler sits in front of every AI provider and enforces the policies your team defines — token quotas, prompt filtering, model fallbacks — in one place.

  • AI Proxy — Route requests to OpenAI, Anthropic, Mistral, or any local LLM through a single endpoint
  • Prompt Templates — Define reusable, version-controlled prompt structures at the gateway layer
  • Prompt Protection — Block injection attacks, PII leakage, and policy-violating inputs before they reach the model
  • Request/Response Transformer — Rewrite AI requests and responses on the fly — model switching without code changes

AI Cost & Usage Control

Track token usage, request counts, and costs across every AI provider — per consumer, per route, per team. Get real-time visibility into your AI spend without changing your application.

Token Quotas

Set per-consumer and per-route token limits. Block overspend before it hits your bill.

Model Routing

Route different request types to different models. Use cheaper models for simple tasks.

Injection Defense

Block prompt injection attacks and policy-violating inputs before they reach the model.

Real-Time Analytics

Track AI usage through Grafana dashboards. Per-model, per-consumer, per-route breakdowns.

Govern Your AI Infrastructure Today

Deploy on-premise. Control every AI call. Pay only for what your team needs.