Energy Considerations of Large Language Model Inference and Efficiency Optimizations
Reviewed by Pithpith:2JM7XEXTopen to challenge →
read the original abstract
As large language models (LLMs) scale in size and adoption, their computational and environmental costs continue to rise. Prior benchmarking efforts have primarily focused on latency reduction in idealized settings, often overlooking the diverse real-world inference workloads that shape energy use. In this work, we systematically analyze the energy implications of common inference efficiency optimizations across diverse Natural Language Processing (NLP) and generative Artificial Intelligence (AI) workloads, including conversational AI and code generation. We introduce a modeling approach that approximates real-world LLM workflows through a binning strategy for input-output token distributions and batch size variations. Our empirical analysis spans software frameworks, decoding strategies, GPU architectures, online and offline serving settings, and model parallelism configurations. We show that the effectiveness of inference optimizations is highly sensitive to workload geometry, software stack, and hardware accelerators, demonstrating that naive energy estimates based on FLOPs or theoretical GPU utilization significantly underestimate real-world energy consumption. Our findings reveal that the proper application of relevant inference efficiency optimizations can reduce total energy use by up to 73% from unoptimized baselines. These insights provide a foundation for sustainable LLM deployment and inform energy-efficient design strategies for future AI infrastructure.
This paper has not been read by Pith yet.
Forward citations
Cited by 6 Pith papers
-
SEFORA: Student Essays with Feedback Corpus and LLM Feedback Evaluation Framework
Releases SEFORA corpus of instructor feedback on college writing and UniMatch evaluation showing no LLM configuration exceeds 0.4 F1 in matching instructor priorities.
-
Pimp My LLM: Leveraging Variability Modeling to Tune Inference Hyperparameters
Variability modeling from software engineering enables systematic sampling, measurement, and prediction of LLM inference configurations for energy, latency, and accuracy trade-offs.
-
SweetSpot: An Analytical Model for Predicting Energy Efficiency of LLM Inference
SweetSpot is an analytical model from Transformer computational and memory complexity that identifies energy minima at short-to-moderate inputs and medium outputs, achieving 1.79% MAPE on H100 GPU measurements across ...
-
Identifying unique developers in OSS projects: A family of models
Creates a ground-truth dataset via LLM-assisted matching and benchmarks classical ML models against an Indel baseline for identifying duplicate developer identities in OSS commit data.
-
From Perception to Action: Can UI Interventions Foster Sustainable LLM Chatbot
A baseline survey of 77 users and a 5-day field study with 11 participants found that sustainability UI interventions increased selection of energy-efficient chatbot modes to 55.8% of prompts.
-
From Cradle to Cloud: A Life Cycle Review of AI's Environmental Footprint
A review of AI sustainability studies finds inconsistent life cycle definitions and predominant reliance on coarse CO2e proxies, with limited coverage of water, materials, and multi-impact assessments.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.