lilbee
Local AI stack in one executable that runs models and searches files, code and web with citations.
Tested on Linux
Use it
Linux x86_64
You need first
- Runtimes:
Python 3.11 to 3.14 - At least 8 GB of RAM
npm install -g lilbee
npm install -g lilbee
Check that it works
which lilbee && lilbee --version 2>&1The output should contain “0.6.90b691”.
First thing to try
lilbee self-check 2>&1Run lilbee self-check first to confirm the tool works on your system.
What it does
- Run and manage local AI modelsTested
How to use it
- Run
lilbee self-checkto confirm the tool is working. - Run
lilbee model listto see which models are installed.
- Run
- Ask a search engine and get cited answersNot tested
- Crawl websites into your local libraryNot tested
- Serve coding agents as an MCP serverNot tested
How far pickfit got
Linux x86_64
- Installs
- Starts
- 1 of 4 features work
- Needs the network
macOS and Windows: the project says it runs there; pickfit hasn’t tried it.
Good for
- search files, notes, code and web with cited answers
- crawl websites into local library
- run coding agents on local models
An alternative to Ollama and LM Studio.
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
Its main function works
- Replayed from scratch
- Needed the network
- Checked on 30 Oct 2026
- Version 0.6.90b691
Differences the test machine hit
- README: npm install -g lilbee then lilbee self-check
In the test: first lilbee invocation downloads ~398MB standalone binary to ~/.cache/lilbee-npm and then runs - README: lilbee add files then ask
In the test: after pulling models, run use-embedder to set embedding_model before add; chat model still unconfigured so ingest/ask needs more setup