Extract entities from a PDF file

Download a PDF file

wget -O "2501_01104.pdf" "https://arxiv.org/pdf/2501.01104"

Download model

wget "https://huggingface.co/onnx-community/gliner_small-v2.1/resolve/main/tokenizer.json"
wget "https://huggingface.co/onnx-community/gliner_small-v2.1/resolve/main/onnx/model.onnx"

Run

Dev

RUST_LOG=debug cargo run -- \
  --pdf-files files.txt \
  --tokenizer-file tokenizer.json \
  --model-file model.onnx \
  --entities technology,organization

Prod

cargo build --release

cp target/release/pdf-ner .

./pdf-ner \
  --pdf-files files.txt \
  --tokenizer-file tokenizer.json \
  --model-file model.onnx \
  --entities conference,name