I am currently working on a web application that uses face-api.js for face detection and recognition. The model files (manifest and shard files) are hosted in the static folder on the server. Every time the application loads, it fetches these files from the server, which increases the loading time of the model.

Is there a way to save these model files (manifest and shard files) in browser storage (like IndexedDB, LocalStorage, or Cache Storage) or directly on the client’s machine to reduce the loading time?

Any guidance on how to implement this or alternative approaches to improve the model loading time.