Dependencies & Setup¶

# 1. Clone repository
git clone <repo-url>
cd Local-File-Organizer

# 2. Install Ollama and pull models
ollama pull qwen2.5:3b-instruct-q4_K_M    # Text: ~1.9 GB
ollama pull qwen2.5vl:7b-q4_K_M           # Vision: ~6.0 GB

# 3. Create virtual environment
python3 -m venv venv
source venv/bin/activate

# 4. Install package
pip install -e .

# 5. Verify
file-organizer --version
fo --version

pip install -e ".[audio]"       # Audio transcription (faster-whisper, torch)
pip install -e ".[video]"       # Video processing (opencv, scenedetect)
pip install -e ".[dedup]"       # Image deduplication (imagededup)
pip install -e ".[archive]"     # Archive support (7z, RAR)
pip install -e ".[scientific]"  # Scientific formats (HDF5, NetCDF, MATLAB)
pip install -e ".[cloud]"       # OpenAI-compatible API providers (openai)
pip install -e ".[cad]"         # CAD formats (ezdxf)
pip install -e ".[desktop]"     # Native desktop window (pywebview + uvicorn)
pip install -e ".[build]"       # Executable packaging (PyInstaller)
pip install -e ".[llama]"       # llama.cpp provider — direct GGUF inference (no Ollama server)
pip install -e ".[mlx]"         # Apple Silicon MLX provider (Darwin only)
pip install -e ".[claude]"      # Anthropic Claude API provider (Claude 3.x text and vision)
pip install -e ".[search]"      # Hybrid BM25 + vector search (rank-bm25, scikit-learn)
pip install -e ".[all]"         # Everything

# pyproject.toml
[project.scripts]
file-organizer = "file_organizer.cli:main"
fo = "file_organizer.cli:main"