地理空间基础模型（Geospatial Foundation Models, GFMs）的基准测试日益依赖聚合得分进行模型排序，但此类排名掩盖了模型差异的成因：性能差距在多大程度上源于架构设计、解码器容量，又在多大程度上源于特定用例的偶然因素？本研究通过受控实验，对欧洲航天局（ESA）Φ-lab 下开发的两款设计理念迥异的 GFM——THOR 与 TerraMind——展开系统性比较。THOR 采用计算自适应架构，支持可变图像块（patch）尺寸，并在原始分辨率下统一融合 Sentinel-1、Sentinel-2 与 Sentinel-3 数据；TerraMind 则是一种多模态生成式 GFM，以双尺度 token/pixel 目标进行预训练，支持任意模态到任意模态的跨模态生成（Thinking-in-Modalities），可在推理阶段推断缺失传感器数据。本研究未报告单一排行榜，而是考察二者在十个涵盖分割与回归任务的多样化应用场景（包括气候灾害响应、甲烷泄漏检测、积雪监测及海冰制图）中，沿若干关键维度的实际差异：图像块尺寸、解码器复杂度、微调策略、输入模态及模型规模。结果表明：架构设计选择——尤其是图像块尺寸与解码器类型——所解释的性能方差大于模型身份本身；两类模型体现了互补的投资策略（TerraMind 侧重预训练阶段的模型规模，THOR 侧重推理阶段的 token 化灵活性）；正确解读结果需以数据集层级的特征刻画为前提。最终呈现的并非单一优胜者，而是一组可迁移的假设及一种诊断性消融方法论，预期可推广至 THOR 与 TerraMind 之外的未来 GFM。

Benchmarks for Geospatial Foundation Models (GFMs) increasingly rank models by aggregate score, but such rankings obscure why models differ: how much of the gap is architecture, how much is decoder capacity, and how much is a use-case-specific artefact? This study addresses that gap through a controlled comparison of two GFMs developed under European Space Agency's $Φ$-lab with contrasting design philosophies: THOR, which introduces a compute-adaptive architecture supporting variable patch sizes and unifies Sentinel-1, -2, and -3 data at their native resolutions; and TerraMind, a multimodal generative GFM pretrained with a dual-scale token/pixel objective that enables any-to-any cross-modal generation (Thinking-in-Modalities) to infer missing sensors at inference time. Rather than reporting a single leaderboard, we investigate the axes along which the two architectures actually differ - patch size, decoder complexity, finetuning regime, input modality, and model scale - across ten use cases spanning segmentation and regression in diverse domains, including climate disaster response, methane leak detection, snow monitoring, or sea ice mapping. We find that architectural design choices - patch size and decoder type in particular - explain more performance variance than model identity itself, that the two models embody complementary investment strategies (pretraining-time scale for TerraMind versus inference-time tokenisation for THOR), and that correctly interpreting results requires dataset-level characterisation. The resulting picture is not a single winner but a set of hypotheses and a diagnostic ablation methodology that we expect to generalise to future GFMs beyond THOR and TerraMind.