Skip to main content

AI Router product description

The AI Router is a unified access service for generative AI models that provides a model catalog, centralized access key management, flexible quotas, consolidated reporting, and end-to-end analytics. With it, you can issue a universal API key for all available models.

The terms of use for the AI Router product are defined in the document Terms of use for individual services: "AI Router" and the SLA.

Use cases

  • connect LLM models from OpenAI, Anthropic, and others via a single API key;
  • use a single code base for every OpenAI-compatible provider;
  • pay for models from a single balance instead of separate bills for each provider;
  • configuring budget limits;
  • monitor consumption via detailed, real-time analytics.

How it works

The AI Router acts as a universal gateway between your application and external LLM providers. The service sends requests via an OpenAI-compatible API to connected models, see the official OpenAI Developers documentation for details. You do not need to configure integrations with each provider separately—using the AI Router API key in your requests is sufficient.

How to use AI Router

To get started, create an AI Router; for more details, see the AI Router: Quick Start.

After creating an AI Router, you can:

Available models

You can view the current list of models after creating an AI Router in the Control Panel: in the top menu click Products → AI Router → AI Router page → tab Models.

Areas of responsibility

Selectel provides

  • access to models via a public API compatible with the OpenAI API;
  • a monitoring system for AI Routers in the Control Panel;
  • consumption analytics;
  • the ability to manage limits for each AI Router and API key;
  • technical support.

Selectel is not responsible

  • for the performance of requests made with API keys, or the results of those requests;
  • for the processing of personal data within the service;
  • for the integration of models into your projects;
  • for the business logic of model operation in your projects.

You are responsible for protecting API keys from compromise.

Pricing

The AI Router is billed according to the pay-as-you-go model. Fees are deducted from your balance hourly based on the volume of model usage for the previous hour; for more information, see the guide AI Router payment model and pricing.

What's included in the price

Unlimited number of tokens for using available models.

Limitations

The AI Router does not support:

  • training AI models;
  • uploading custom models to the catalog;
  • deploying models from the catalog on-premises, in a data center, on dedicated servers, or in a certified cloud.