Introduction

Routify streamlines requests to 150+ open & closed source models with a unified API and easy-to-use UI. It is also supports caching, fallbacks, retries, timeouts, loadbalancing, and can be self-hosted.

• Fast: Routify is built on top of .NET8 and is designed to be fast and efficient.
• Easy to use: Routify provides an easy-to-use interface where you can manage all requests and responses.
• Schema mapping: Routify can map requests and responses to different provider schemas.
• Observability: Routify provides observability features to help you monitor your AI requests.
• Cost control: Routify can help you control costs by estimating the cost of each request and applying cost limits.
• Load balancing: Routify can load balance requests across multiple providers.
• Failover: Routify can automatically failover to a different provider if the primary provider is down.

How it works

Routify uses routes which allow you to have different behavior and logic for different use cases. Each route is defined by a path and can have one or more providers to proxy the requests.

For example, you can have a route that matches /openai to proxy requests to the OpenAI API and another route that matches /anthropic to proxy requests to the Anthropic API.

Each route has a schema that defines which provider schema to use for the route. This will determine the structure of the request and response objects. You can use the OpenAI schema for input and output but have the request proxied to the Anthropic API or any other provider. This allows you to switch providers without changing your application code.

Also you can override request parameters, such as system prompt, temperature, max_tokens, top_p, etc. for each provider, through the user interface, without changing your application code.