Cesium Sandcastle adds AI-driven semantic search as generally available

Cesium Sandcastle has made its AI-driven semantic search generally available. The company shared progress on the feature in December 2025. Sandcastle has long been a cornerstone of the CesiumJS developer community. It supports showcasing new features, sharing ideas, learning, prototyping, experimenting, and validating code. Its accessibility allows developers to quickly find, build, and share code examples relevant to their projects, with demonstrations on how to implement features in the CesiumJS library. Users can now perform a wider range of searches with the semantic search to more efficiently find sandcastles relevant to their queries. They no longer need to exactly match the name, description, or code to find applicable examples. The feature is powered by an open-source embedding model, which embeds the user's query and compares it to precomputed semantic embeddings for the full code contents of all sandcastles. The model runs locally on the end user's hardware. Search data does not leave the device. The semantic search functions as a hybrid with the previous search, so existing results continue to appear as before. Users can disable the Semantic Search toggle in settings to opt out. Feedback can be shared on GitHub.