You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
PDF extraction, image analysis, and single-document benchmarking CLI. convert turns PDFs into Markdown, analyze inspects images/charts, and benchmark runs one PDF across multiple providers with per-tool outputs and results.json.
Directory Structure
Path
Purpose
Key Files
pdftoolkit/
Main package
cli.py, benchmark.py, clients.py, utils.py
pdftoolkit/providers/
Provider adapters
convert.py, analyze.py
eval/
Optional research/eval runners
run_eval.py, tools/*.py
src/
Reference scripts
*.py
tests/
Test suite
test_cli.py, test_benchmark.py, test_*
docs/
Sample PDFs
*.pdf
output/
Generated outputs
benchmark/, *.md
Entry Points
Entry
File
Description
CLI
pdftoolkit/cli.py
Typer app for convert, benchmark, and analyze
Benchmark registry
pdftoolkit/benchmark.py
Tool metadata, default selection, JSON result writing