Skip to main content

Documentation Index

Fetch the complete documentation index at: https://dev.docs.inworld.ai/llms.txt

Use this file to discover all available pages before exploring further.

Realtime Router is an intelligent routing layer that helps you select the right model and configuration for your use case, to maximize the performance and user metrics you care about from cost and latency to user retention and revenue. In addition to providing a unified API to access hundreds of LLMs through a single endpoint while automatically handling fallbacks, Realtime Router enables you to easily run A/B experiments, route different user segments to different models, and measure the impact on your KPIs. This means you can actually optimize for your specific application and your specific users.

Developer quickstart

Learn how to make your first API call in minutes with a guided tutorial.

Core concepts

Understand the core concepts behind Realtime Router
Realtime Router is currently in research preview. Please share any feedback with us in Discord.

​
Key Benefits

  • Unified API: Access models from OpenAI, Anthropic, Google, and more through a single API
  • High reliability: Automatically fall back to other providers if one fails
  • Dynamic selection: Optimize the model or provider in real-time based on price, speed, or intelligence
  • Cost optimization: Automatically choose the most cost-effective provider or model for each request to help you stay within budget
  • Live experimentation: Easily run experiments on different models and prompts to see what works best for your users
  • Insightful analytics: Seamlessly integrate with your metrics to understand how different models impact your KPIs