User Documentation

Getting started with NativeLab.

Everything you need to install, configure, and use NativeLab for local LLM inference.

Getting Started

Install and first run.

NativeLab requires Python 3.10+ and installs via pip. The GUI auto-downloads llama.cpp binaries on first launch.

Quick Install

$ pip install nativelab
$ nativelab        # launches GUI
$ nativelab --cli  # terminal mode

First Launch

  • Auto-detects hardware (CPU, RAM, GPU)
  • Downloads llama.cpp runtime
  • Recommends a model for your device
  • Opens the chat workspace

Guides

User documentation.

Step-by-step guides for common tasks.

Installation guide Terminal CLI Feature catalogue GUI tour and theming Troubleshooting
Workflows and pipelines Model registry Labs experimentation Integrations PhonoLab (Android)

Quick Links

  • Chat with a local model
  • Build a visual pipeline
  • Attach documents to chat
  • Use API models
  • Connect Android phone
  • Set up Ollama bridge

PhonoLab

NativeLab on Android.

The official Android client for local LLM inference on your phone.

PhonoLab Docs

Architecture, API endpoints, error handling, device discovery, and build instructions.

PhonoLab page

API Reference

Complete endpoint documentation for PhonoLab's LAN API server. curl examples, vision, RAG, config.

API endpoints