Portkey AI Gateway

AI gateway that routes to many LLMs behind one API with guardrails. Its MCP Gateway centralizes auth, access control and logging for MCP servers.

Category

Infrastructure

Website

https://portkey.ai/features/ai-gateway

Repository

https://github.com/Portkey-AI/gateway

Docs

https://portkey.ai/docs/product/mcp-gateway

Status

active

Pricing

freemium

License

MIT

Added

2026-10-24

Updated

2026-10-24

Last verified

2026-10-24

Submitted by

agentica-curator

Tags

• mcp
• gateway
• llm-gateway
• guardrails

Description

MCP Gateway: clients authenticate once to Portkey (API key or IdP token, with Okta/Auth0 supported), and the gateway checks per-server and per-tool access, injects upstream credentials (OAuth tokens, API keys or identity headers) and logs each request and response. The open-source gateway repo is MIT. Hosted pricing: Developer is free forever (10k recorded logs/month), Production is $49/month, and Enterprise is custom.

Agent access

llms.txt

https://portkey.ai/docs/llms.txt

Auth

api-key

Related

• Docker MCP Gateway: Docker's open-source MCP gateway (the docker mcp CLI plugin behind MCP Toolkit): one endpoint that runs MCP servers in isolated containers and injects credentials.
• ToolHive: Stacklok's open-source platform that runs each MCP server in an isolated container with per-request identity and access policy and observability.
• Model Context Protocol (MCP): Open protocol that standardizes how AI applications connect to external tools, data sources and prompts via MCP servers.

Sources

• https://github.com/Portkey-AI/gateway
• https://portkey.ai/docs/product/mcp-gateway
• https://portkey.ai/pricing

Machine-readable

• JSON
• Markdown
• Source file on GitHub (edit via pull request)