sist2app/sist2 is a self-hosted file search engine that indexes the files on your own computer and then helps you find them again. The project name is short for Simple incremental search tool. The most recent release the repository has made public is v4.2.1, but the material available to verify does not state what changed in this version.
The core of this tool is extracting and indexing metadata, body text, and even text read out by OCR from many kinds of files. It provides full-text search and previews over a collection of local files. The README describes it as a file search tool with low memory usage that runs multi-threaded, and it offers text and metadata extraction, thumbnail generation, OCR support, and a mobile-friendly web interface. Incremental scanning — re-scanning only newly added or changed files to update the index — is also built in by default.
You can choose between Elasticsearch and SQLite for the search backend, and in a Docker environment you can manage and schedule scan jobs through the web interface. The primary language shown on GitHub is C and the license is GPL-3.0, with 1323 stars; the repository was created on September 28, 2019. It stands out that a project begun in 2019 is still putting out releases.
That is all that can be confirmed. There is only the README's feature list and the repository's basic information, and no independent measurements showing how fast it is or up to what scale it has been validated. The v4.2.1 changelog is missing as well, so to know how the new version differs from the previous one you have to check the release notes directly. Since the basis does not go beyond what is written in the documents, rather than something to place large performance expectations on, it is closer to a candidate worth checking in an environment where you need an index over your own collection of files.