TensorOpera Router: A Multi-Model Router for Efficient LLM Inference

TL;DR

TensorOpera Router (TO-Router) is a multi-model system that dynamically routes queries to different LLMs, improving efficiency and reducing costs while maintaining or improving performance.

Contribution

We introduce TO-Router, a system that integrates multiple LLMs and intelligently routes queries to optimize efficiency, cost, and performance.

Findings

Up to 40% improvement in query efficiency

Cost reductions of up to 30%

Performance maintained or improved by up to 10%

Abstract

With the rapid growth of Large Language Models (LLMs) across various domains, numerous new LLMs have emerged, each possessing domain-specific expertise. This proliferation has highlighted the need for quick, high-quality, and cost-effective LLM query response methods. Yet, no single LLM exists to efficiently balance this trilemma. Some models are powerful but extremely costly, while others are fast and inexpensive but qualitatively inferior. To address this challenge, we present TO-Router, a non-monolithic LLM querying system that seamlessly integrates various LLM experts into a single query interface and dynamically routes incoming queries to the most high-performant expert based on query's requirements. Through extensive experiments, we demonstrate that when compared to standalone expert models, TO-Router improves query efficiency by up to 40\%, and leads to significant cost reductions of up to 30%, while maintaining or enhancing model performance by up to 10%.