Routoo: Learning to Route to Large Language Models Effectively
This addresses the cost-efficiency problem for LLM users by providing a method to reduce inference costs without compromising quality, though it appears incremental as it builds on existing routing concepts.
The paper tackles the problem of balancing quality and cost in LLM deployment by introducing Routoo, an architecture that optimizes LLM selection for specific prompts. Results show Routoo matches Mixtral 8x7b performance with one-third lower inference costs, and with GPT4 in the pool, it nearly matches GPT4 performance at half the cost.
LLMs with superior response quality--particularly larger or closed-source models--often come with higher inference costs, making their deployment inefficient and costly. Meanwhile, developing foundational LLMs from scratch is becoming increasingly resource-intensive and impractical for many applications. To address the challenge of balancing quality and cost, we introduce Routoo, an architecture designed to optimize the selection of LLMs for specific prompts based on performance, cost, and efficiency. Routoo provides controllability over the trade-off between inference cost and quality, enabling significant reductions in inference costs for a given quality requirement. Routoo comprises two key components: a performance predictor and cost-aware selector. The performance predictor is a lightweight LLM that estimates the expected performance of various underlying LLMs on a given prompt without executing them. The cost-aware selector module then selects the most suitable model based on these predictions and constraints such as cost and latency, significantly reducing inference costs for the same quality. We evaluated Routoo using the MMLU benchmark across 57 domains employing open-source models. Our results show that Routoo matches the performance of the Mixtral 8x7b model while reducing inference costs by one-third. Additionally, by allowing increased costs, Routoo surpasses Mixtral's accuracy by over 5% at equivalent costs, achieving an accuracy of 75.9%. When integrating GPT4 into our model pool, Routoo nearly matches GPT4's performance at half the cost and exceeds it with a 25% cost reduction. These outcomes highlight Routoo's potential to significantly reduce inference costs without compromising quality, and even to establish new state-of-the-art results by leveraging the collective capabilities of multiple LLMs.