The field of geospatial reasoning is rapidly advancing with the application of large language models (LLMs). Recent research has focused on improving the ability of LLMs to understand and reason about geometric and spatial relationships. This has led to the development of new frameworks and benchmarks for evaluating the performance of LLMs on geospatial tasks. Notably, the use of neuro-symbolic approaches and multimodal synthesis has shown promise in enhancing the spatial perception and reasoning abilities of LLMs. Furthermore, the integration of LLMs with other technologies, such as robot arms, has opened up new avenues for creative applications in fields like dance improvisation. Overall, the field is moving towards the development of more advanced and specialized models that can effectively reason about complex geospatial relationships. Noteworthy papers include: Foundation Models for Geospatial Reasoning, which demonstrates the potential of LLMs in understanding geometries and topological spatial relations. NeSyGeo, which proposes a novel neuro-symbolic framework for generating geometric reasoning data and achieves significant improvements in the performance of multiple MLLMs. GeoGramBench, which establishes a valuable benchmark for evaluating the geometric program reasoning capabilities of LLMs. AutoGPS, which presents a neuro-symbolic collaborative framework for automated geometry problem solving with concise and reliable reasoning processes.