Build requirements

• autoconf automake ca-certificates clang cmake dpkg-dev g++ gcc git golang-go libc6-dev libssl-dev make python3 python3-dev python3-pip python3-venv rsync unzip wget
• nodejs 20

Poetry

pipx install --force poetry

qdrant

Common

protobuf-compiler

Debian 13

rustup

Debian 12

wget -O ~/.local/bin/rustup-init https://static.rust-lang.org/rustup/dist/x86_64-unknown-linux-gnu/rustup-init
chmod +x ~/.local/bin/rustup-init
rustup-init

Runtime requirements

• libgl1 libmagic1

Clone and install

git clone --recursive ...
cd ...
./install

Using the configurator

Start

.venv/bin/python -m learn2rag

Usage

Set up a language model

Go to “Models”, add a new model with an OpenAI-compatible API URL like https://example.com/api, a token like sk-mytoken, and a model like gemma-3-27b-it.

Set up sources

Go to “Sources”, add your sources by specifying a path to the directory with the files to import.

Set up a pipeline

Go to “Pipelines”, add a new pipeline and select a storage path for the pipeline data, your configured language model and sources.

Import documents

Use “Start” → “import”. Wait until the pipeline is done running.

Run the pipeline and UI

Use “Start” → “pipeline”. The UI will be available at http://localhost:5001/ with default credentials user@learn2rag.de and learn2rag.

See also

• compose
• ui

Building a package

Requirements

• pyenv with relevant Python versions

Building

./bundle-linux

Running a package

Requirements

• unzip

Running

unzip learn2rag.zip
cd bundle
./firstrun.sh
./configurator