smartpdfplumber 0.1.2

A wrapper around LangChain’s PDFPlumber integration with added support for image-aware data extraction.

Smart PDF Plumber

A lightweight wrapper over LangChain’s PDFPlumber integration that extends PDF parsing with image understanding—extracting not just text, but also contextual insights from embedded images.

It is designed for two common cases:

• Extract plain text from PDFs page by page.
• Optionally describe embedded images and include those descriptions in the page text.

Features

• Page-level PDF parsing with pdfplumber.
• Optional character deduplication for PDFs with repeated text.
• Optional image description support using either Google Gemini or Hugging Face vision-language models.
• LangChain-friendly output: a list of Document objects with metadata such as source path, page number, and total pages.

Installation

Install from PyPI:

pip install smartpdfplumber

The project currently targets Python 3.13 or newer.

Quick Start

from SmartPDFPlumber.smart_pdfplumber import SmartPDFLoader

loader = SmartPDFLoader("path/to/file.pdf")
documents = loader.load()

for document in documents:
	print(document.metadata)
	print(document.page_content)

Options

SmartPDFLoader forwards keyword arguments to PDFPlumberParser:

SmartPDFLoader(
	file_path="path/to/file.pdf",
	text_kwargs={"x_tolerance": 2},
	dedupe=True,
	describe_image=False,
	model=None,
)

text_kwargs

Extra keyword arguments passed to pdfplumber.Page.extract_text().

dedupe

Set dedupe=True to call page.dedupe_chars() before extracting text. This can help with PDFs that repeat characters in the output.

describe_image

Set describe_image=True to include image descriptions inline in the page text.

When this is enabled, you must also provide model.

model

Supported values:

• gemini
• huggingface

Image Descriptions

Google Gemini

from SmartPDFPlumber.smart_pdfplumber import SmartPDFLoader

loader = SmartPDFLoader(
	"path/to/file.pdf",
	describe_image=True,
	model="gemini",
)
documents = loader.load()

This path uses google-genai. Make sure your Google authentication is configured before running it.

Hugging Face

from SmartPDFPlumber.smart_pdfplumber import SmartPDFLoader

loader = SmartPDFLoader(
	"path/to/file.pdf",
	dedupe=True,
	describe_image=True,
	model="huggingface",
)
documents = loader.load()

This path uses transformers, torch, and torchvision to run a vision-language model locally.

Output Format

Each page becomes a LangChain Document with metadata similar to:

{
	"source": "path/to/file.pdf",
	"file_path": "path/to/file.pdf",
	"page": 0,
	"total_pages": 12,
}

The page value is zero-based.

Example

from SmartPDFPlumber.smart_pdfplumber import SmartPDFLoader

loader = SmartPDFLoader(
	"assets/sample.pdf",
	dedupe=True,
	describe_image=True,
	model="huggingface",
)

documents = loader.load()

for document in documents:
	print(document.page_content[:200])

Notes

• If you enable image descriptions without passing model, the parser raises a ValueError.
• If you use model="gemini", google-genai must be installed and authenticated.
• If you use model="huggingface", the model is loaded lazily and cached for reuse.

License

MIT