lilbee: a single-executable local search engine you can talk to, with its HTTP API served by Litestar #4853
tobocop2
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Thanks for Litestar. It runs lilbee's REST/HTTP API. lilbee is a local-first search engine you can talk to: it runs and manages its own local models, indexes your files and code, crawls the websites you point it at, and answers with a citation to the source, all in one executable that exposes a CLI, a Textual TUI, an MCP server, a Python library, and this HTTP server. GUIs build on the HTTP server: the Obsidian plugin one, and it talks to lilbee over it.
Some cool details: it's the same single process, not a sidecar. The Litestar app shares the search engine, the embeddings, and the model subprocesses with the rest of lilbee, so the HTTP API answers from the exact same index the TUI does, no separate service to stand up. The models run in-process today, and I'm moving them to a llama-server and llama-swap architecture so the same binary scales across multiple GPUs while staying one executable (tobocop2/lilbee#267). There's a published API reference at https://lilbee.sh/api/.
Site: https://lilbee.sh.
Repo: https://github.com/tobocop2/lilbee
Obsidian plugin: https://obsidian.lilbee.sh
Beta Was this translation helpful? Give feedback.
All reactions