Kong AI Gateway
API gatewaykong-ai-gateway
Kong Gateway AI plugins that proxy, load-balance, cache and guard traffic to LLM providers.
- Provider
- default
- Kind
- API gateway
- Category
- Edge
| property | type | values | required | kind default | this component |
|---|---|---|---|---|---|
| auth | enum | none | api-key | jwt | oauth2 | mtls | no | none | inherited |
| balancerAlgorithm | enum | round-robin | consistent-hashing | least-connections | lowest-latency | lowest-usage | priority | semantic | no | round-robin | inherited |
| balancerRetries | integer | — | no | 5 | inherited |
| concurrencyLimit | boolean | — | no | false | inherited |
| idempotencyKeys | boolean | — | no | false | inherited |
| instances | integer | — | no | — | inherited |
| keyTtlHours | integer | — | no | — | inherited |
| managed | boolean | — | no | false | ▸ false |
| promptGuard | boolean | — | no | false | inherited |
| protocol | enum | http | grpc | websocket | fix | ouch | no | http | ▸ http |
| rateLimiting | boolean | — | no | false | inherited |
| semanticCache | boolean | — | no | false | inherited |
| tls | boolean | — | no | true | inherited |
Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.
sync-chain-too-deep
Every consecutive synchronous hop from this entry point spends part of the stated p95 latency budget, and the chain's failures multiply. More than three hops is a budget nobody is tracking.
- — all of these hold:
- — the design requirement `latency.p95Ms` has a value
- — the longest chain of synchronous calls starting at the node is more than 3 hops long
nodes:
- { id: "kong-ai-gateway", component: "kong-ai-gateway" }API Gatewayalibaba-api-gatewayAmazon API Gatewayaws-api-gatewayAWS AppSync GraphQL APIaws-appsyncAzure API Managementazure-api-managementAI Gatewaycloudflare-ai-gatewayCloudflare API Shieldcloudflare-api-shieldMCP Server Portalcloudflare-mcp-portalCloudflare Privacy Gatewaycloudflare-privacy-gatewayGeneric API gatewayapi-gatewayApache APISIXapisixApollo GraphOSapollo-graphosEasegresseasegressEmissary-ingressemissaryEnvoy Gatewayenvoy-gatewayGloo GatewayglooHasurahasuraHelicone AI Gatewayhelicone-gatewayHigresshigressKong GatewaykongKong Konnectkong-konnectKrakenDkrakendLiteLLM ProxylitellmMartian GatewaymartianNot Diamondnot-diamondPortkey AI GatewayportkeyRequestyrequestySpring Cloud Gatewayspring-cloud-gatewayTyk GatewaytykUnifyunify-aiVercel AI Gatewayvercel-ai-gatewayWSO2 API Managerwso2-api-managerZuulzuulApigeeapigeeAgent Gatewaygoogle-agent-gatewayAPI Gatewaygoogle-api-gatewayApigee hybridgoogle-apigee-hybridCloud Endpointsgoogle-cloud-endpointsFirebase Data Connectgoogle-firebase-data-connectIBM API Connectibm-api-connectOCI API Gatewayoci-api-gateway
# yaml-language-server: $schema=../../schema/component.schema.json # Kong AI Gateway: Kong Gateway plugins (AI Proxy, AI Proxy Advanced, AI Semantic Cache, AI Prompt Guard) # that front LLM providers, with the data plane self-hosted or Konnect-managed. Per "AI Proxy Advanced # Plugin Configuration Reference", the balancer defaults to round-robin with 5 retries and fails over only # on error and timeout; per "AI Semantic Cache Plugin", caching is a separate Enterprise plugin you enable. schemaVersion: "1" type: "component" id: "kong-ai-gateway" kind: "api-gateway" provider: "default" docs: "https://developer.konghq.com/plugins/ai-proxy-advanced/reference/" defaults: managed: false protocol: "http" properties: balancerAlgorithm: type: "enum" values: ["round-robin", "consistent-hashing", "least-connections", "lowest-latency", "lowest-usage", "priority", "semantic"] default: "round-robin" balancerRetries: type: "integer" default: 5 semanticCache: type: "boolean" default: false promptGuard: type: "boolean" default: false