Portkey
AI gateway and observability platform for routing, monitoring, and governing LLM apps
Introduction
Portkey is a control panel for AI applications. Its AI Gateway routes requests across many LLMs through one endpoint, and it adds prompt management, guardrails, and an observability suite for cost, quality, and latency. It works with agent frameworks such as LangChain, CrewAI, and AutoGen.
- Added on
- Nov 2, 2026
- Website
- portkey.ai
- Company
- Portkey, Inc.
Product information
What is Portkey?
Portkey is an AI platform for teams building LLM applications. It provides an AI gateway, prompt management, guardrails, and monitoring in one place.
How to use Portkey?
Point your app's OpenAI-compatible API base URL at the Portkey endpoint. Requests are then routed and logged, and you can manage prompts and parameters in one place.
Core features of Portkey
- AI Gateway for LLM routing
- Prompt management
- Guardrails for LLM behavior
- Observability for cost, quality, and latency
- MCP client for agents with tool access
Use cases of Portkey
- 1Monitoring cost, quality, and latency of AI apps
- 2Routing requests to many LLMs through one endpoint
- 3Managing prompts across a team
- 4Enforcing guardrails on LLM output
Frequently asked questions about Portkey
How does Portkey work?+
You replace the API base path in your app with the Portkey endpoint, and Portkey routes requests to the model provider while giving you visibility and control.
Will Portkey slow down my app?+
Portkey states that it benchmarks added latency closely and uses caching, failover, and edge layers that can improve response times.
Which category is Portkey in?+
Portkey is listed under AI Code Editor.