Jun 13Vibe with Hermes Agent — Bengaluru · RSVP
ToolsMCPBlogResearchCommunityStar on GitHub
All Tools
K
Dev ToolsFreemiumOpen Source

KONG

The cloud-native API and AI gateway

Apache-2.0

ABOUT

Teams building AI-powered applications need to route requests to multiple LLM providers (OpenAI, Anthropic, Google, self-hosted models) while managing API keys, rate limits, usage costs, and failover — but each provider has its own auth scheme, rate limits, and pricing. Kong provides a unified AI gateway layer that proxies LLM requests, enforces usage policies, handles provider failover, and collects telemetry across all providers from a single control plane.

INSTALL
docker run -d --name kong --network host kong:latest

INTEGRATION GUIDE

1. Route LLM inference requests across multiple providers with automatic failover and load balancing 2. Enforce API key authentication, rate limits, and usage quotas for internal and external AI API consumers 3. Collect unified metrics and logs for all LLM API traffic across OpenAI, Anthropic, self-hosted models, and more 4. Implement semantic caching of LLM responses to reduce costs and improve latency at the gateway level 5. Secure AI endpoints with prompt injection detection, PII redaction, and content moderation policies in the request path

TAGS

api-gatewayai-gatewayllm-routingapi-managementcloud-nativekubernetesmicroservices
Kong — AI Tool | Agentic AI For Good