Installation

Basic install

The base install gives you the Python library, CLI, and semantic search. No Docker required.

pip install locallens

This includes:

• LocalLens Python API
• locallens CLI
• Qdrant Edge (embedded vector database)
• Sentence-transformers embeddings (all-MiniLM-L6-v2)
• Extractors for PDF, DOCX, code, text, spreadsheets

With LLM support (RAG Q&A)

The ask command requires Ollama running locally. Install Ollama separately, then pull a model:

# Install Ollama from https://ollama.com
ollama pull qwen2.5:3b
ollama serve

No extra Python packages needed — LocalLens talks to Ollama over HTTP.

With voice

Adds speech-to-text (Moonshine) and text-to-speech (Piper) for the voice interface:

pip install "locallens[voice]"

With MCP server

Exposes LocalLens as tools for Claude Code, Claude Desktop, Cursor, and other MCP clients:

pip install "locallens[mcp]"

With web dashboard

Full React web UI with FastAPI backend:

pip install "locallens[server]"

With Rust acceleration

Installs locallens-core, the optional Rust extension package for 5-10x faster BM25, chunking, and file walking on large corpora:

pip install "locallens[fast]"

Without this, LocalLens uses pure Python for everything. Same results, just slower on 10k+ file corpora.

Everything

Install all optional dependencies at once (includes Rust extensions):

pip install "locallens[all]"

From source

git clone https://github.com/mahimairaja/locallens.git
cd locallens
pip install -e ".[all]"

System requirements

Requirement	Minimum	Recommended
Python	3.11+	3.12+
RAM	2 GB	4 GB
OS	macOS, Linux, Windows	macOS (Apple Silicon)
Docker	Not required	For web dashboard only
Ollama	Not required	For ask / RAG feature