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

TL;DR

LLMSpace is a pioneering framework that models the carbon footprint of running large language models on solar-powered LEO satellites, considering operational and embodied emissions.

Contribution

It introduces the first comprehensive carbon modeling framework for LLM inference on AI-enabled LEO satellites, accounting for hardware, workload, and lifecycle factors.

Findings

Reveals trade-offs between carbon footprint, latency, and hardware design.

Highlights the significance of satellite lifecycle emissions in total carbon impact.

Provides insights for designing sustainable space-based LLM inference systems.

Abstract

Large language models (LLMs) impose rapidly growing energy demands, creating an emerging energy and carbon crisis driven by large-scale inference. Solar-powered, AI-enabled low Earth orbit (LEO) satellites have been proposed to mitigate terrestrial electricity consumption, but their lifecycle carbon footprint remains poorly understood due to launch emissions, satellite manufacturing, and radiation-hardened hardware requirements. This paper presents \textit{LLMSpace}, the first carbon modeling framework for LLM inference on AI-enabled LEO satellites. LLMSpace jointly models operational and embodied carbon, peripheral subsystems, radiation-hardened accelerators and memories, and LLM-specific workload characteristics such as prefill-decode behavior and token generation. Using realistic satellite and GPU configurations, LLMSpace reveals key trade-offs among carbon footprint, inference latency, hardware design, and operational lifetime for sustainable space-based LLM inference. Source code: https://github.com/UnchartedRLab/LLMSpace.