I'm building lilbee, the whole local AI stack in a single executable: it runs and manages the models, and it's a search engine you can talk to. It indexes your files, notes, and code, crawls web pages into your library, and every answer cites the exact file and line. It started as me trying to build an Encarta 99 out of my own stuff that I could talk to privately and terminal first, and that's still the mission.

Why lilbee

Onboarding with local AI is painful. A typical stack is a model server, a vector database, a document ingestion pipeline, a web UI, and a Docker Compose file holding it all together. Every piece needs configuring before you can ask a single question.

lilbee is that whole stack in one self-contained executable: model manager, search engine over your own data, MCP for agents, HTTP for GUIs, and a TUI for the terminal. No containers, no networking, nothing else to install. It scales from one laptop to as many GPUs as you have.

I'm trying to democratize local AI, starting with talking to your own stuff, while building the tool I want for myself.

What I'm building

lilbee

Terminal-first, because I built it for myself and I knew that if the architecture stayed flexible, friendlier surfaces would come easily. TUI, CLI, REST API, MCP server, and Python library, with multi-GPU placement and integrations for opencode and hermes so programmers can put local AI behind their favorite agents. lilbee is its own model manager: you don't need Ollama or LM Studio at all, and the model families it runs are tested end to end on real GPUs, the architectures behind most of the 190,000+ GGUF models on Hugging Face. Already running Ollama or LM Studio? lilbee connects to them and keeps your models.

obsidian-lilbee

One of those friendlier surfaces, the gateway for Q&A with local models. My favorite story here is a completely non-technical user who went from the plugin store to talking to his solar panel docs in minutes, without ever opening a terminal.

xberg

Extraction is the core of lilbee. You can't search what you can't read, and an answer is only as good as the text that came out of the document, so a scanned PDF that turns to mush or a table that loses its columns is a bad answer waiting to happen. Good extraction is what makes good search possible.

In early iterations of lilbee I built my own extraction layer. When I found xberg I gutted all of that work, and a lot of it became my earliest contributions there. Today I'm a maintainer across the xberg ecosystem, and most of my open source work goes there (all my merged PRs):

Project	Stars	Contributions
xberg		Merged PRs
crawlberg		Merged PRs
html-to-markdown		Merged PRs
liter-llm		Merged PRs
alef		Merged PRs
tree-sitter-language-pack		Merged PRs
sceptre		Merged PRs

Upstream work

lilbee is built on a lot of open source projects. Stuff I've contributed outside xberg that I'm pretty proud of:

Project	Stars	What I did
DBOS Transact (Python, Go)		Made SQLite migrations safe when several processes open a new database, in Python and Go.
pdf_oxide		Filed dozens of reproducible bug reports and landed fixes for text extraction, rendering crashes, and nondeterminism.
lancedb / lance		Made Lance run on older processors, with compatible builds and a source package.
GPUStack / gguf-parser-go		Reworked the VRAM estimator GPUStack uses to size models, including low estimates that silently ran out of memory.
huggingface_hub / xet-core		Fixed a download crash in headless programs.
llama-cpp-python		Fixed models that wouldn't load because of their chat template.
ik_llama.cpp		Made the server report its memory use per device, so programs don't have to scrape logs. A rewrite of my llama.cpp change for this fork.
SkyPilot		Four open PRs that help the API server survive a dying process, back off after errors, and cap its log.
llama.cpp		Made llama-server report its memory use per device, so programs don't have to scrape logs. The port to ik_llama.cpp is merged.
candle		Fixed DeBERTa-v2 models crashing at half precision.
safetensors		Fixed large tensors failing to load on 32-bit targets like wasm.

Misc throughout the years

Project	Stars	What I did
mercurius		Fixed a crash in apps with nested loaders.
screenshot-glb		Added a timeout so slow renders can finish.
MetalAPI		My first project, from college: the Metal Archives scraped and served as a REST API.
go-utils		Added unique and fixed reverse in Goldziher's Go library, years before he founded xberg and we ended up working together.
simpler-cpp-json		Automatic JSON serialization for any C++ struct, years before nlohmann/json shipped the same idea.
openapi-generator		Unity support for the C# generator.

Elsewhere

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|t|o|b|i|a|s|@|l|i|l|b|e|e|.|s|h|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+