Quick-start Guide

This short guide walks you through converting a document to VXDF, inspecting it, and querying it from Python.

1. Install VXDF

pip install vxdf

2. Convert a PDF via the CLI

python -m vxdf convert sample.pdf sample.vxdf

If this is the first time you use an OpenAI embedding model, the CLI will prompt for an API key and store it in ~/.vxdf/config.toml.

3. Inspect the file

python -m vxdf info sample.vxdf

4. Query from Python

from vxdf.reader import VXDFReader

with VXDFReader("sample.vxdf") as reader:
    print(reader.header)
    chunk = reader.get_chunk("doc_0")
    print(chunk["text"])

That’s it! See the vxdf for the full API.