GreenServ: Energy-Efficient Context-Aware Dynamic Routing for Multi-Model LLM Inference

TL;DR

GreenServ is a dynamic, context-aware routing framework for multi-model LLM inference that optimizes accuracy and energy efficiency by adaptively selecting models based on query features using a multi-armed bandit approach.

Contribution

It introduces a novel adaptive routing method for LLM inference that learns online without extensive offline calibration, improving efficiency and accuracy.

Findings

GreenServ outperforms static and random routing baselines.

Achieved 22% higher accuracy and 31% energy reduction compared to random routing.

Attained an average accuracy of 71.7% on benchmark tasks.

Abstract

Large language models (LLMs) demonstrate remarkable capabilities, but their broad deployment is limited by significant computational resource demands, particularly energy consumption during inference. Static, one-model-fits-all inference strategies are often inefficient, as they do not exploit the diverse range of available models or adapt to varying query requirements. This paper presents GreenServ, a dynamic, context-aware routing framework that optimizes the trade-off between inference accuracy and energy efficiency. GreenServ extracts lightweight contextual features from each query, including task type, semantic cluster, and text complexity, and routes queries to the most suitable model from a heterogeneous pool, based on observed accuracy and energy usage. We employ a multi-armed bandit approach to learn adaptive routing policies online. This approach operates under partial feedback, eliminates the need for extensive offline calibration, and streamlines the integration of new models into the inference pipeline. We evaluated GreenServ across five benchmark tasks and a pool of 16 contemporary open-access LLMs. Experimental results show that GreenServ consistently outperforms static (single-model) and random baselines. In particular, compared to random routing, GreenServ achieved a 22% increase in accuracy while reducing cumulative energy consumption by 31%. Finally, we evaluated GreenServ with RouterBench, achieving an average accuracy of 71.7% with a peak accuracy of 75.7%. All artifacts are open-source and available here: \href{https://github.com/TZData1/llm-inference-router}{GitHub}