# The Real Cost of Context: Predicting Compute Time and Energy Consumption in LLM Prefill and Decode Phases

DevFeed: [The Real Cost of Context: Predicting Compute Time and Energy Consumption in LLM Prefill and Decode Phases](<https://devfeed.tech/articles/the-real-cost-of-context-predicting-compute-time-and-energy-consumption-in-llm-prefill-and-decode-phases-108217.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.39621>)

Author: Lars Krupp, Daniel Gei{\\ss}ler, Francisco M. Calatrava-Nicolas, Stefan Gerd Fritsch, Paul Lukowicz, Jakob Karolus

Published: 2026-10-30T04:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Sustainable Computing](<https://devfeed.tech/topics/sustainable-computing.md>), [context window](<https://devfeed.tech/topics/context-window.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-workflows](<https://devfeed.tech/tags/agentic-workflows.md>), [ai](<https://devfeed.tech/tags/ai.md>), [compute](<https://devfeed.tech/tags/compute.md>), [context](<https://devfeed.tech/tags/context.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [decode](<https://devfeed.tech/tags/decode.md>), [energy](<https://devfeed.tech/tags/energy.md>), [llm](<https://devfeed.tech/tags/llm.md>), [prefill](<https://devfeed.tech/tags/prefill.md>), [time](<https://devfeed.tech/tags/time.md>)

## AI overview

The study finds that LLM decode compute time and energy scale linearly with output length, while prefill time and energy show both linear and quadratic scaling with input length. This means large input contexts can materially affect resource use in agentic workloads, an effect that output-token-focused estimates may miss. The authors report that compute time can serve as a proxy for energy consumption when direct measurements are unavailable, and present a mapping framework for estimating energy. The supplied text does not specify the profiled models, measurement setup, or framework's validation details.

## Source excerpt

arXiv:2610.39621v1 Announce Type: new Abstract: The integration of Large Language Models (LLMs) into agentic architectures has substantially increased token consumption, raising concerns about their energy footprint. Yet, estimating the energy consumption of proprietary API-based models remains challenging, and existing approaches focus primarily on generated output tokens while overlooking the impact of the large input contexts characteristic of agentic workflows. To address this gap, we systematically profile compute time and energy consumption across a wide range of prefill (input) and decode (output) token lengths. Using generalized linear models, we show that decode compute time and energy scale linearly with output length, whereas prefill compute time and energy exhibit both linear and quadratic scaling with input length. These findings demonstrate that input context becomes a relevant factor for large-context workloads. Furthermore, we show that compute time serves as a reliable proxy for energy consumption when direct energy measurements are unavailable. Our mapping framework enables precise energy estimation and supports more sustainable deployment decisions for agentic AI.