FAQ

Is my data sent to the cloud?

No. All processing happens locally on your machine. Documents are parsed, chunked, and embedded using Ollama running on localhost. There is no telemetry, no analytics, and no tracking. The only external network call is downloading Ollama itself (one-time, user-initiated).

What AI models can I use?

Locke works with any model available through Ollama, including Llama 3.1, Mistral, Gemma, and more. You can switch models at any time from the Settings page.

What about images?

Locke supports an optional vision model that can extract text from images and photos. This lets you index screenshots, scanned documents, and diagrams alongside your other files.

How much disk space do I need?

Plan for approximately 10 GB of free disk space. This covers the AI models (~4-8 GB depending on the model) and the vector database that stores your document embeddings.

How much RAM do I need?

8 GB minimum, 16 GB recommended. Larger models like Llama 3.1 8B run best with 16 GB or more.

What file formats are supported?

Locke supports 43 file types including PDF, DOCX, XLSX, PPTX, CSV, HTML, images, Markdown, JSON, and many code file types. Maximum file size is 100 MB per document.

Do documents update automatically?

Yes. Locke watches your document folders and re-indexes files automatically when they change.

Can I use Locke without documents?

By default, Locke only answers questions grounded in your documents. You can enable Allow General Chat in Settings to let the AI respond using its own knowledge when no documents match.

What platforms are supported?

Locke runs on Windows 10+, macOS 12+ (Intel and Apple Silicon), and Linux (64-bit).

How much does Locke cost?

Locke is a one-time purchase of $29.99. No subscriptions, no recurring fees. Lifetime updates are included. Download it from the releases page.

Can I customize the appearance?

Yes. Locke ships with 4 color themes: Cyan, Violet, Rose, and Amber.