Requirements

Ollama — Locke will install it for you on first launch, or download it from ollama.com
8 GB RAM minimum (16 GB recommended)
~10 GB free disk space for AI models and vector database
Windows 10+, macOS 12+, or Linux (64-bit)

Download

Go to the Releases page and download the installer for your platform:

Windows: Locke-*-win-x64.exe
macOS (Intel): Locke-*-mac-x64.dmg
macOS (Apple Silicon): Locke-*-mac-arm64.dmg
Linux: Locke-*-linux-x64.AppImage or .deb

First Launch

The setup wizard will guide you through:

Installing Ollama (if not already installed)
Downloading an AI model (default: Llama 3.1 8B)
Adding your first document folder

Adding Documents

Open the Settings page
Click Add Folder
Select a folder containing your documents
Locke will automatically parse, chunk, and embed all supported files

Asking Questions

Once your documents are indexed, go to the Chat page and start typing. Locke retrieves the most relevant chunks from your documents and generates an answer with source citations. Click a source citation to preview the original file.

Supported File Types

Locke supports 43 file types including:

PDF — .pdf
Word — .docx, .doc
Spreadsheets — .xlsx, .xls
Presentations — .pptx
CSV — .csv
HTML — .html, .htm
Images — with optional vision model for extracting text from images and photos
Text & Config — .txt, .md, .json, .log, .yaml, .yml, .xml, .ini, .cfg, .conf, .toml
Code — .py, .js, .ts, .java, .c, .cpp, .h, .go, .rs, .rb, .sh, .bat, .ps1, .sql, .r

Maximum file size: 100 MB per document.