Choosing to Be Green: Advancing Green AI via Dynamic Model Selection

2025-09-24

AI Summary

The high energy consumption of AI models—particularly large language models—exacerbates environmental impact, necessitating energy-efficiency optimization under strict accuracy constraints. Method: This paper proposes a Green AI Dynamic Model Selection framework that innovatively integrates model cascading and intelligent routing mechanisms. It jointly models energy-efficiency estimation, accuracy constraints, and task-specific requirements to enable adaptive selection of optimal submodels during inference. Contribution/Results: Empirically validated on real-world datasets, the framework reduces energy consumption by approximately 25% compared to a single high-energy model while maintaining over 95% prediction accuracy. It provides a scalable hybrid strategy and practical paradigm for building adaptive, energy-efficient AI systems that co-optimize accuracy and energy efficiency.

Technology Category

• Machine Learning: Efficient ML / Green AI
• Search and Optimization: Learning to Search
• Computer Vision: Large Vision Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Artificial Intelligence is increasingly pervasive across domains, with ever more complex models delivering impressive predictive performance. This fast technological advancement however comes at a concerning environmental cost, with state-of-the-art models - particularly deep neural networks and large language models - requiring substantial computational resources and energy. In this work, we present the intuition of Green AI dynamic model selection, an approach based on dynamic model selection that aims at reducing the environmental footprint of AI by selecting the most sustainable model while minimizing potential accuracy loss. Specifically, our approach takes into account the inference task, the environmental sustainability of available models, and accuracy requirements to dynamically choose the most suitable model. Our approach presents two different methods, namely Green AI dynamic model cascading and Green AI dynamic model routing. We demonstrate the effectiveness of our approach via a proof of concept empirical example based on a real-world dataset. Our results show that Green AI dynamic model selection can achieve substantial energy savings (up to ~25%) while substantially retaining the accuracy of the most energy greedy solution (up to ~95%). As conclusion, our preliminary findings highlight the potential that hybrid, adaptive model selection strategies withhold to mitigate the energy demands of modern AI systems without significantly compromising accuracy requirements.

Problem

• Reducing AI environmental footprint via dynamic model selection
• Balancing computational efficiency with predictive accuracy requirements
• Mitigating energy demands of complex models without compromising performance

Innovation

• Dynamic model selection for reducing AI environmental footprint
• Green AI cascading and routing methods for energy efficiency
• Hybrid adaptive strategies balancing accuracy and sustainability