The Cost and Network Limits of Space-Based AI Compute
Not on the map. Why?
Abstract (arXiv)
This paper evaluates whether large-scale AI data centers deployed in low-Earth orbit (LEO) could become a cost-effective alternative to terrestrial facilities. The analysis compares orbital and ground-based systems across launch cost, power generation, cooling, radiation exposure, and atmospheric reentry, as well as compute-network performance. A key distinction is the shift from terrestrial Clos networks to space-based mesh networks using laser inter-satellite links. Using bisection bandwidth, bisection intensity, and roofline-style models, we show that while LEO-based inference may be feasible, training frontier-scale LLMs in orbit is unlikely to be competitive with terrestrial data centers.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 9 of its 31 references that are in the map. First posted on arXiv 15 Jul 2026.
Builds on
9 of 31 references.
References · 9
- 2024The Llama 3 Herd of Modelscited by 14,715
- 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
- 2022Efficiently Scaling Transformer Inferencecited by 539
- 2022Reducing Activation Recomputation in Large Transformer Modelscited by 302
- 2024MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUscited by 112
- 2025WLB-LLM: Workload-Balanced 4D Parallelism for Large Language Model Trainingcited by 17
- 2024Performance Modeling and Workload Analysis of Distributed Large Language Model Training and Inferencecited by 9
- 2025Towards a future space-based, highly scalable AI infrastructure system designcited by 4
- 2026Evaluating Cross-Architecture Performance Modeling of Distributed ML Workloads Using StableHLOcited by 1
Built on by
Not cited yet.