AI Gateway · JSON Structure

Ai Gateway Route Structure

A single model route on an AI gateway, binding a client-facing alias and matchers to upstream provider backends with caching, fallback, fanout, rate limit, budget, and guardrail policy.

Type: object Properties: 10 Required: 2

AI GatewayLLM RouterLLM ProxyModel RoutingPrompt FirewallGuardrailsAI ObservabilityCost ControlsAI GovernanceAPI Gateway

AIGatewayRoute is a JSON Structure definition published by AI Gateway, describing 10 properties, of which 2 are required. It conforms to the https://json-structure.org/meta/core/v0/# meta-schema.

Properties

name modelAlias matchers backends strategy cache rateLimit budget guardrails tenant

Meta-schema: https://json-structure.org/meta/core/v0/#

JSON Structure

{
  "$schema": "https://json-structure.org/meta/core/v0/#",
  "$id": "https://raw.githubusercontent.com/api-evangelist/ai-gateway/refs/heads/main/json-structure/ai-gateway-route-structure.json",
  "name": "AIGatewayRoute",
  "description": "A single model route on an AI gateway, binding a client-facing alias and matchers to upstream provider backends with caching, fallback, fanout, rate limit, budget, and guardrail policy.",
  "type": "object",
  "properties": {
    "name": {
      "type": "string",
      "description": "Unique route name within the gateway.",
      "example": "primary-llm"
    },
    "modelAlias": {
      "type": "string",
      "description": "Client-facing model alias the gateway exposes.",
      "example": "gpt-4o-primary"
    },
    "matchers": {
      "type": "array",
      "description": "Inbound matchers selecting this route.",
      "items": {
        "type": "object",
        "properties": {
          "type": { "type": "string", "example": "header" },
          "key": { "type": "string", "example": "x-model-alias" },
          "value": { "type": "string", "example": "primary-llm" }
        }
      }
    },
    "backends": {
      "type": "array",
      "description": "One or more upstream provider backends serving this route.",
      "items": {
        "type": "object",
        "properties": {
          "provider": { "type": "string", "example": "openai" },
          "model": { "type": "string", "example": "gpt-4o-2024-08-06" },
          "weight": { "type": "int32", "example": 100 },
          "priority": { "type": "int32", "example": 1 },
          "credentialRef": { "type": "string", "example": "openai-prod" }
        }
      }
    },
    "strategy": {
      "type": "string",
      "description": "How the gateway selects among multiple backends.",
      "example": "fallback"
    },
    "cache": {
      "type": "object",
      "properties": {
        "mode": { "type": "string", "example": "semantic" },
        "ttlSeconds": { "type": "int32", "example": 3600 },
        "similarityThreshold": { "type": "float", "example": 0.95 }
      }
    },
    "rateLimit": {
      "type": "object",
      "properties": {
        "requestsPerMinute": { "type": "int32", "example": 600 },
        "tokensPerMinute": { "type": "int32", "example": 200000 },
        "concurrency": { "type": "int32", "example": 32 }
      }
    },
    "budget": {
      "type": "object",
      "properties": {
        "currency": { "type": "string", "example": "USD" },
        "limit": { "type": "float", "example": 500.0 },
        "period": { "type": "string", "example": "monthly" },
        "action": { "type": "string", "example": "block" }
      }
    },
    "guardrails": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": { "type": "string", "example": "pii-redaction" },
          "stage": { "type": "string", "example": "both" },
          "policyRef": { "type": "string", "example": "pii-redaction-default" }
        }
      }
    },
    "tenant": {
      "type": "string",
      "description": "Logical tenant owning this route.",
      "example": "platform"
    }
  },
  "required": ["name", "backends"]
}