You can take a look at txtai (https://github.com/neuml/txtai). It can run in a Python process. It has support for storing content in SQLite and embeddings vectors in local vector index formats (Faiss, HNSW, Annoy).

Disclaimer: I'm the primary author of txtai.

I actually just finished a POC using DuckDB that does similarity search for HN comments.

https://github.com/patricktrainer/hackernews-comment-search

FWIW, Simon Willison's `llm` tool just uses SQLite plus a few UDFs. The simplicity of that approach is appealing to me but I don't have a good sense of when+why it becomes insufficient.

For Python I believe Chroma [1] can be used embedded.

For Go I recently started building chromem-go, inspired by the Chroma interface: https://github.com/philippgille/chromem-go

It's neither advanced nor for scale yet, but the RAG demo works.

[1] https://github.com/chroma-core/chroma

The qdrant clients support a local mode where you point to a file[0].

[0] -- https://github.com/qdrant/qdrant-client#local-mode

I've used usearch successfully for a small project: https://github.com/unum-cloud/usearch/

Qdrant as a library project https://github.com/tyrchen/qdrant-lib