Pith. sign in

REVIEW 11 cited by

Quantifying the Energy Consumption and Carbon Emissions of LLM Inference via Simulations

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2507.11417 v1 pith:RMNB2P56 submitted 2025-07-15 cs.DC

Quantifying the Energy Consumption and Carbon Emissions of LLM Inference via Simulations

Miray \"Ozcan , Philipp Wiesner , Philipp Wei{\ss} , Odej Kao This is my paper
classification cs.DC
keywords inferencecarbonemissionsenergypowersimulationanalysiscarbon-aware
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

The environmental impact of Large Language Models (LLMs) is rising significantly, with inference now accounting for more than half of their total lifecycle carbon emissions. However, existing simulation frameworks, which are increasingly used to determine efficient LLM deployments, lack any concept of power and, therefore, cannot accurately estimate inference-related emissions. We present a simulation framework to assess the energy and carbon implications of LLM inference under varying deployment setups. First, we extend a high-fidelity LLM inference simulator with a GPU power model that estimates power consumption based on utilization metrics, enabling analysis across configurations like batch size, sequence length, and model parallelism. Second, we integrate simulation outputs into an energy system co-simulation environment to quantify carbon emissions under specific grid conditions and explore the potential of carbon-aware scheduling. Through scenario-based analysis, our framework reveals how inference parameters affect energy demand and carbon footprint, demonstrates a renewable offset potential of up to 69.2% in an illustrative deployment case, and provides a foundation for future carbon-aware inference infrastructure design.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. LLMSpace: Carbon Footprint Modeling for Large Language Model Inference on LEO Satellites

    cs.LG 2026-05 unverdicted novelty 7.0

    LLMSpace is the first modeling framework that jointly calculates operational and embodied carbon emissions for LLM inference on LEO satellites, incorporating radiation-hardened hardware, peripheral systems, and LLM wo...

  2. LLMSpace: Carbon Footprint Modeling for Large Language Model Inference on LEO Satellites

    cs.LG 2026-05 unverdicted novelty 7.0

    LLMSpace is the first framework to jointly model operational and embodied carbon for LLM inference on LEO satellites, incorporating radiation-hardened hardware, peripheral systems, and workload patterns such as prefil...

  3. Evaluating the Environmental Impact of using SLMs and Prompt Engineering for Code Generation

    cs.SE 2026-04 unverdicted novelty 7.0

    Chain-of-Thought prompting balances high accuracy with low energy use in small language models for code generation, while multi-sampling strategies add high energy costs for small accuracy gains.

  4. WattGPU: Predicting Inference Power and Latency on Unseen GPUs and LLMs

    cs.DC 2026-07 unverdicted novelty 6.0

    WattGPU ML models predict LLM inference power and latency on unseen GPUs with median errors of 3.4-13.5% using public data and show better performance than baselines.

  5. EnergyLens: Interpretable Closed-Form Energy Models for Multimodal LLM Inference Serving

    cs.CV 2026-05 unverdicted novelty 6.0

    EnergyLens derives a twelve-parameter closed-form energy model via symbolic regression that achieves 88.2% top-1 configuration accuracy with 50 samples and extrapolates to unseen batch sizes and hardware.

  6. EnergyLens: Interpretable Closed-Form Energy Models for Multimodal LLM Inference Serving

    cs.CV 2026-05 unverdicted novelty 6.0

    EnergyLens derives a 12-parameter closed-form energy model for multimodal LLM inference serving that fits from ~50 samples, achieves 88% top-1 configuration accuracy, and extrapolates to new batch sizes and hardware.

  7. Tackling the Data-Parallel Load Balancing Bottleneck in LLM Serving: Practical Online Routing at Scale

    cs.DC 2026-05 unverdicted novelty 6.0

    BalanceRoute uses a piecewise-linear F-score (with optional short lookahead) for sticky request routing in LLM serving, reducing DP imbalance and raising end-to-end throughput versus vLLM baselines on production and A...

  8. Tackling the Data-Parallel Load Balancing Bottleneck in LLM Serving: Practical Online Routing at Scale

    cs.DC 2026-05 unverdicted novelty 6.0

    BalanceRoute reduces data-parallel imbalance in LLM inference via F-score routing and lookahead, yielding higher end-to-end throughput on 144-NPU clusters versus vLLM baselines.

  9. KAIROS: Stateful, Context-Aware Power-Efficient Agentic Inference Serving

    cs.DC 2026-04 unverdicted novelty 6.0

    KAIROS reduces power by 27% on average (up to 39.8%) for agentic AI inference by using long-lived context to jointly manage GPU frequency, concurrency, and request routing across instances.

  10. LLM Harms: A Taxonomy and Discussion

    cs.CY 2025-12 reject novelty 3.0

    Proposes a five-bucket taxonomy of LLM harms and calls for dynamic auditing, but the systematic review behind it is not reproducible and contains mismatched citations.

  11. LLM Harms: A Taxonomy and Discussion

    cs.CY 2025-12 unverdicted novelty 3.0

    This paper proposes a taxonomy of LLM harms in five categories and suggests mitigation strategies plus a dynamic auditing system for responsible development.