Llama-Scan: Convert PDFs to Text W Local LLMs
6 days ago
- #multimodal-models
- #Ollama
- #PDF-to-text
- Convert PDFs to text files locally with no token costs using Ollama.
- Utilize the latest multimodal models from Ollama for detailed text descriptions of images and diagrams.
- Requires Python 3.10+ and Ollama installed and running locally.
- Install the default model with `ollama run qwen2.5vl:latest`.
- Install the tool via pip (`pip install llama-scan`) or uv (`uv tool install llama-scan`).
- Basic usage: `llama-scan path/to/your/file.pdf` with options for output directory, model selection, image handling, and page range.
- Example command for processing specific pages with custom width: `llama-scan document.pdf --start 1 --end 5 --width 1000`.
- Option to use a different Ollama model: `llama-scan document.pdf --model qwen2.5vl:3b`.