SpaceMoE: Towards Orbital General Intelligence with Distributed Mixture-of-Experts Inference
This work provides a conceptual framework for enabling AGI in satellite networks, but it is a position paper without experimental results, making it an incremental contribution.
SpaceMoE proposes a distributed mixture-of-experts inference framework for deploying large language models on satellites, addressing onboard memory, computation, and energy constraints. The paper reviews industrial progress, introduces the architecture, and discusses key design problems like expert placement and routing, highlighting satellite-specific challenges.
As satellite networks evolve to support increasingly diverse services and artificial general intelligence (AGI), large language models (LLMs) are emerging as a critical foundation for future space systems. However, deploying LLMs on satellites is hindered by stringent constraints on onboard memory, computation, and energy. In this context, the mixture-of-experts (MoE) architecture emerges as a promising solution, leveraging sparse expert activation to enable scalable model inference. By harnessing the architectural advantages of MoE, this article provides a comprehensive overview of SpaceMoE, a new paradigm for distributed MoE inference in satellite networks. We first review recent industrial progress and emerging standardization trends that motivate the evolution toward space AGI systems. Then, we introduce the fundamentals and architectural evolution of SpaceMoE. Subsequently, we discuss three fundamental design problems in SpaceMoE, namely expert placement, expert selection, and hidden-state transmission and routing, highlighting how satellite-specific factors such as dynamic topology, battery degradation, and thermal limits fundamentally reshape their solutions. Finally, we outline promising research directions for realizing scalable, efficient, and sustainable on-orbit MoE inference in future satellite networks.