Geospatial Foundation Models (GFMs) provide powerful representations, but high compute costs hinder their widespread use. Pre-computed embedding data products offer a practical "frozen" alternative, yet they currently exist in a fragmented ecosystem of incompatible formats and resolutions. This lack of standardization creates an engineering bottleneck that prevents meaningful model comparison and reproducibility. We formalize this landscape through a three-layer taxonomy: Data, Tools, and Value. We survey existing products to identify interoperability barriers. To bridge this gap, we extend TorchGeo with a unified API that standardizes the loading and querying of diverse embedding products. By treating embeddings as first-class geospatial datasets, we decouple downstream analysis from model-specific engineering, providing a roadmap for more transparent and accessible Earth observation workflows.

翻译：地理空间基础模型（GFMs）提供了强大的表征能力，但其高昂的计算成本阻碍了广泛应用。预计算的嵌入数据产品提供了一种实用的"冻结"替代方案，然而目前它们存在于格式与分辨率互不兼容的碎片化生态系统中。这种标准化缺失造成了工程瓶颈，阻碍了有意义的模型比较与可复现性研究。我们通过三层分类体系（数据层、工具层、价值层）对这一领域进行形式化梳理，并系统调研现有产品以识别互操作性障碍。为弥合这一鸿沟，我们扩展了TorchGeo框架，提供统一API以标准化各类嵌入产品的加载与查询。通过将嵌入数据视为一等地理空间数据集，我们将下游分析与模型特定工程解耦，为构建更透明、更可访问的地球观测工作流提供了技术路线图。