Instructions to use PaddlePaddle/PP-DocBee2-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PaddleOCR
How to use PaddlePaddle/PP-DocBee2-3B with PaddleOCR:
# 1. See https://www.paddlepaddle.org.cn/en/install to install paddlepaddle # 2. pip install paddleocr from paddleocr import DocVLM model = DocVLM(model_name="PP-DocBee2-3B") output = model.predict( input={"image": "path/to/image.png", "query": "Parsing this image and output the content in Markdown format."}, batch_size=1 ) for res in output: res.print() res.save_to_json(save_path="./output/res.json") - Notebooks
- Google Colab
- Kaggle
Download model_state.pdparams from PaddlePaddle/PP-DocBee2-3B: direct link, hf CLI and curl.
- Browser
- Download file 8.13 GB
-
https://huggingface.co/PaddlePaddle/PP-DocBee2-3B/resolve/main/model_state.pdparams
- Command line
-
hf download hf://PaddlePaddle/PP-DocBee2-3B/model_state.pdparams
-
curl -L -o model_state.pdparams https://huggingface.co/PaddlePaddle/PP-DocBee2-3B/resolve/main/model_state.pdparams
8.13 GB
- Xet hash:
- c6a3ca36e9921ce18132a0b916d636b1cf0b149b7f65215633666faf1fad78ff
- Size of remote file:
- 8.13 GB
- SHA256:
- f2c661f1d71de128c190333e818fd8d9395812818ea14e3c5c36409b812b86da
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.