api-gateway

Kong/kong

🦍 The API and AI Gateway

songquanpeng/one-api

A unified OpenAI-style gateway for dozens of LLM providers — One API lets you call OpenAI, Azure, Claude, Gemini, DeepSeek, ChatGLM, Ernie Bot, Qwen and more through a single standard OpenAI API format. Ships as a single binary with a Docker image for one-click deployment, aimed at teams needing centralized key management and resale of model access.

LLM API 管理 & 分发系统，支持 OpenAI、Azure、Anthropic Claude、Google Gemini、DeepSeek、字节豆包、ChatGLM、文心一言、讯飞星火、通义千问、360 智脑、腾讯混元等主流模型，统一 API 适配，可用于 key 管理与二次分发。单可执行文件，提供 Docker 镜像，一键部署，开箱即用。LLM API management & key redistribution system, unifying multiple providers under a single API. Single binary, Docker-ready, with an English UI.

chaitin/SafeLine

SafeLine: a self-hosted WAF that blocks SQLi, XSS and bot traffic — SafeLine is a self-hosted web application firewall and reverse proxy that detects and blocks SQL injection, XSS, HTTP floods, brute-force logins and other common exploits using semantic analysis rather than simple pattern matching. It also offers CAPTCHA and acts as an API gateway, making it a practical option for teams wanting to protect web apps without relying on a cloud WAF service.

SafeLine is a self-hosted WAF(Web Application Firewall) / reverse proxy to protect your web apps from attacks and exploits.

apache/apisix

The Cloud-Native API Gateway and AI Gateway

higress-group/higress

An Envoy-based gateway that turns any API into an MCP server for AI agents — Higress is a cloud-native API gateway built on Istio and Envoy, extensible via Wasm plugins in Go, Rust or JS. Its AI gateway layer supports mainstream model providers and can expose existing APIs as MCP servers, aimed at teams building LLM-powered agent infrastructure alongside traditional API management.

🤖 AI Gateway | AI Native API Gateway

maximhq/bifrost

A zero-config gateway unifying 23+ LLM providers behind one API — Bifrost is a Go-based AI gateway giving one OpenAI-compatible API across 23+ providers, with automatic failover, load balancing, semantic caching, virtual keys and budgets. It deploys in seconds via npx or Docker, includes a web UI for config and monitoring, and adds under 200µs overhead at 5k RPS—aimed at teams building reliable, multi-provider AI apps and agents.

Open-source, high-performance AI gateway that works as your LLM gateway, MCP gateway and agent gateway. One OpenAI-compatible API for 10,000+ models, with failover, load balancing, semantic caching, virtual keys, budgets and MCP Code Mode. Works with any SDK or agent and adds under 200µs of overhead at 5k RPS.

tbphp/gpt-load

A self-hosted gateway for juggling multiple AI API keys and accounts — GPT-Load is a Go-based AI gateway built for teams or individuals managing many API keys and subscription accounts across providers like OpenAI, Gemini, Claude and Codex. It unifies them behind one entry point with traffic scheduling, failover, request logs and usage accounting, useful for anyone tired of juggling scattered credentials manually.

Self-hosted AI gateway for multi-channel, multi-credential setups — API keys and subscription accounts, scheduling, failover, request logs and usage. 自托管 AI 网关：多渠道多凭据统一接入，含密钥与订阅账号、调度容错、日志与用量。

agentgateway/agentgateway

A dedicated proxy layer for MCP and A2A agent traffic — agentgateway is a Rust-based proxy built specifically for AI-native protocols like MCP and A2A, aiming to be a complete connectivity layer for agentic AI systems. It targets teams running AI agents and MCP servers who need gateway, routing, and service-mesh capabilities, including Kubernetes-native deployments via Gateway API.

Next Generation Agentic Proxy for AI Agents and MCP servers

looplj/axonhub

A self-hosted gateway that unifies 100+ LLM APIs behind one endpoint — AxonHub lets any SDK call OpenAI, Anthropic, Gemini, DeepSeek and more without code changes, routing requests through one gateway. It adds failover, load balancing, cost tracking and end-to-end tracing, aimed at teams running agents or apps across multiple LLM providers who want reliability and spend control.

⚡️ Open-source AI Gateway — Use any SDK to call 100+ LLMs. Built-in failover, load balancing, cost control & end-to-end tracing.

theagentrouter/agent-router

Envoy AI Gateway rebrands as Agent Router, a unified control plane — Formerly Envoy AI Gateway, now under the Agentic AI Foundation. It gives app teams one OpenAI-compatible API across hosted and self-hosted models plus MCP servers, while platform teams manage credentials, routing, quotas and failover via Envoy. CRDs, CLI and images stay unchanged, easing migration for existing users.

Manages Unified Access to Generative AI Services built on Envoy Gateway

HelpCode-ai/anythingmcp

Self-hosted gateway that wraps 517 APIs as MCP tools — AnythingMCP connects REST/OpenAPI, SOAP, GraphQL, OData and SQL systems to Claude, ChatGPT and Copilot without writing a custom MCP server. It ships 517 ready-made connectors, including SAP, Etsy and Amazon Seller, with 23 needing no API key, plus RBAC, SSO and audit logs for enterprise use.

MCP gateway for AI agents: turn any REST/OpenAPI, GraphQL, SOAP or SQL API into MCP tools. 562 ready-made connectors (SAP, ERP, e-commerce) with RBAC, SSO, audit logs and a knowledge graph.