Watt Counts: Energy-Aware Benchmark for Sustainable LLM Inference on Heterogeneous GPU Architectures

• URL: http://arxiv.org/abs/2604.09048v1
• Date: Fri, 10 Apr 2026 07:15:58 GMT
• Title: Watt Counts: Energy-Aware Benchmark for Sustainable LLM Inference on Heterogeneous GPU Architectures
• Abstract summary: Watt Counts is the largest open-access dataset of energy consumption of Large Language Models (LLMs)<n>We show that practitioners can reduce energy consumption by up to 70% in server scenarios with negligible impact on user experience.
• Score: 0.9558392439655014
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: While the large energy consumption of Large Language Models (LLMs) is recognized by the community, system operators lack guidance for energy-efficient LLM inference deployments that leverage energy trade-offs of heterogeneous hardware due to a lack of energy-aware benchmarks and data. In this work we address this gap with Watt Counts: the largest open-access dataset of energy consumption of LLMs, with over 5,000 experiments for 50 LLMs across 10 NVIDIA Graphics Processing Units (GPUs) in batch and server scenarios along with a reproducible, open-source benchmark that enables community submissions to expand this dataset. Leveraging this dataset, we conduct a system-level study of LLM inference across heterogeneous GPU architectures and show that GPU selection is crucial for energy efficiency outcomes and that optimal hardware choices vary significantly across models and deployment scenarios, demonstrating the critical importance of hardware-aware deployment in heterogeneous LLM systems. Guided by our data and insights, we show that practitioners can reduce energy consumption by up to 70% in server scenarios with negligible impact on user experience, and by up to 20% in batch scenarios.

Related papers

• Sustainable LLM Inference for Edge AI: Evaluating Quantized LLMs for Energy Efficiency, Output Accuracy, and Inference Latency [6.306413686006502]
  We conduct a comprehensive analysis of 28 quantized Large Language Models (LLMs) from the Ollama library.<n>We evaluate energy efficiency, inference performance, and output accuracy across multiple quantization levels and task types.<n>Our findings reveal the trade-offs between energy efficiency, inference speed, and accuracy in different quantization settings.
  arXiv  Detail & Related papers  (2025-04-04T11:29:30Z)
• WDMoE: Wireless Distributed Mixture of Experts for Large Language Models [68.45482959423323]
  Large Language Models (LLMs) have achieved significant success in various natural language processing tasks. We propose a wireless distributed Mixture of Experts (WDMoE) architecture to enable collaborative deployment of LLMs across edge servers at the base station (BS) and mobile devices in wireless networks.
  arXiv  Detail & Related papers  (2024-11-11T02:48:00Z)
• Mini-InternVL: A Flexible-Transfer Pocket Multimodal Model with 5% Parameters and 90% Performance [78.48606021719206]
  Mini-InternVL is a series of MLLMs with parameters ranging from 1B to 4B, which achieves 90% of the performance with only 5% of the parameters. We develop a unified adaptation framework for Mini-InternVL, which enables our models to transfer and outperform specialized models in downstream tasks.
  arXiv  Detail & Related papers  (2024-10-21T17:58:20Z)
• MLPerf Power: Benchmarking the Energy Efficiency of Machine Learning Systems from Microwatts to Megawatts for Sustainable AI [5.50579824344998]
  Machine learning (ML) technologies have led to a surge in power consumption across diverse systems.<n>This paper introduceserf Power, a comprehensive benchmarking methodology to evaluate the energy efficiency of ML systems at power levels ranging from microwatts to megawatts.
  arXiv  Detail & Related papers  (2024-10-15T20:06:33Z)
• DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency [7.073435885680335]
  We propose DynamoLLM, the first energy-management framework for generative large language models. At a service-level, DynamoLLM conserves 53% energy and 38% operational carbon emissions, and reduces 61% cost to the customer.
  arXiv  Detail & Related papers  (2024-08-01T17:40:45Z)
• Hybrid Heterogeneous Clusters Can Lower the Energy Consumption of LLM Inference Workloads [0.2389598109913753]
  Training and using Large Language Models (LLMs) require large amounts of energy. This paper addresses the challenge of reducing energy consumption in data centers running LLMs. We propose a hybrid data center model that uses a cost-based scheduling framework to dynamically allocate tasks across hardware accelerators.
  arXiv  Detail & Related papers  (2024-04-25T11:24:08Z)
• Distributed Inference and Fine-tuning of Large Language Models Over The Internet [91.00270820533272]
  Large language models (LLMs) are useful in many NLP tasks and become more capable with size. These models require high-end hardware, making them inaccessible to most researchers. We develop fault-tolerant inference algorithms and load-balancing protocols that automatically assign devices to maximize the total system throughput.
  arXiv  Detail & Related papers  (2023-12-13T18:52:49Z)
• Federated Fine-Tuning of LLMs on the Very Edge: The Good, the Bad, the Ugly [62.473245910234304]
  This paper takes a hardware-centric approach to explore how Large Language Models can be brought to modern edge computing systems. We provide a micro-level hardware benchmark, compare the model FLOP utilization to a state-of-the-art data center GPU, and study the network utilization in realistic conditions.
  arXiv  Detail & Related papers  (2023-10-04T20:27:20Z)
• From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inference [19.439683873290623]
  Large language models (LLMs) have exploded in popularity due to their new generative capabilities that go far beyond prior state-of-the-art. These models carry significant computational challenges, especially the compute and energy costs required for inference.
  arXiv  Detail & Related papers  (2023-10-04T17:41:59Z)
• Data-Juicer: A One-Stop Data Processing System for Large Language Models [73.27731037450995]
  A data recipe is a mixture of data from different sources for training Large Language Models (LLMs) We build a new system named Data-Juicer, with which we can efficiently generate diverse data recipes. The data recipes derived with Data-Juicer gain notable improvements on state-of-the-art LLMs.
  arXiv  Detail & Related papers  (2023-09-05T08:22:07Z)
• FusionAI: Decentralized Training and Deploying LLMs with Massive Consumer-Level GPUs [57.12856172329322]
  We envision a decentralized system unlocking the potential vast untapped consumer-level GPU. This system faces critical challenges, including limited CPU and GPU memory, low network bandwidth, the variability of peer and device heterogeneity.
  arXiv  Detail & Related papers  (2023-09-03T13:27:56Z)
• A Framework for Energy and Carbon Footprint Analysis of Distributed and Federated Edge Learning [48.63610479916003]
  This article breaks down and analyzes the main factors that influence the environmental footprint of distributed learning policies. It models both vanilla and decentralized FL policies driven by consensus. Results show that FL allows remarkable end-to-end energy savings (30%-40%) for wireless systems characterized by low bit/Joule efficiency.
  arXiv  Detail & Related papers  (2021-03-18T16:04:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.