Skip to main content
Portrix is a unified AI gateway that sits between your application and the world’s leading AI providers. Instead of maintaining separate API integrations for OpenAI, Anthropic, Google, Mistral, and dozens of others, you connect once to Portrix and get immediate access to more than 400 large language models through a single, OpenAI-compatible endpoint.

​
Why Portrix?

Building production AI applications today means wrestling with a fragmented ecosystem. You juggle multiple SDKs, rotate API keys across providers, manage different request formats and response schemas, and write custom retry logic for every service you rely on. When a provider goes down or raises its prices, the cost of switching is high. Portrix solves this by acting as a universal translation and routing layer. You write your code once against the OpenAI API format, and Portrix handles authentication, protocol conversion, and request routing to whichever model you need. Switching from GPT-4o to Claude 3.5 Sonnet is a one-word change in your code.

​
Key capabilities

Universal Model Access

Access 400+ models from OpenAI, Anthropic, Google, Mistral, Cohere, and more from a single endpoint. The full catalog is available with one API key.

OpenAI Compatible

Portrix implements the OpenAI API specification. Any SDK or tool that works with OpenAI — LangChain, LlamaIndex, Vercel AI SDK, and others — works with Portrix out of the box.

Smart Routing & Fallbacks

Configure automatic fallbacks so requests retry against a secondary model if the primary provider is unavailable. Load balancing distributes traffic across providers to manage rate limits.

Managed Web Search

Enable real-time web search for any model through Portrix’s managed search tool, so your agents and assistants always have access to up-to-date information.

​
What you can build

Portrix is a general-purpose gateway suitable for a wide range of AI-powered products:
  • Chatbots and assistants — Build conversational interfaces that can switch between models to balance cost, speed, and capability per conversation.
  • Autonomous agents — Give your agents access to every frontier model plus managed web search, all through a single integration.
  • Search-augmented AI — Combine retrieval with generation using Portrix’s built-in web search support, without managing a separate search API.
  • Cost-optimized pipelines — Route simple tasks to smaller, cheaper models and reserve expensive frontier models for complex reasoning, all within the same codebase.

​
Next steps

Quickstart

Follow the step-by-step guide to send your first request through Portrix in under 5 minutes.

API Reference

Dive into the full API reference to explore every endpoint and parameter available in Portrix.