ai-models/PaddleOCR-VL
2
0
Fork 0
mirror of https://www.modelscope.cn/PaddlePaddle/PaddleOCR-VL.git synced 2026-04-02 21:42:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

update readme

Browse Source
This commit is contained in:
TingquanGao
2025-10-17 08:24:02 +00:00
parent c320f18247
commit 78c582ff62
1 changed files with 21 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
README.md
View File

@ -3,7 +3,6 @@ license: apache-2.0
pipeline_tag: image-text-to-text pipeline_tag: image-text-to-text
tags: tags:
- ERNIE4.5 - ERNIE4.5
- PaddleOCR
- PaddlePaddle - PaddlePaddle
- image-to-text - image-to-text
- ocr - ocr
@ -17,7 +16,6 @@ language:
- en - en
- zh - zh
- multilingual - multilingual
library_name: PaddleOCR
--- ---
<div align="center"> <div align="center">
@ -39,7 +37,7 @@ PaddleOCR-VL: Boosting Multilingual Document Parsing via a 0.9B Ultra-Compact Vi
[![License](https://img.shields.io/badge/license-Apache_2.0-green)](./LICENSE) [![License](https://img.shields.io/badge/license-Apache_2.0-green)](./LICENSE)
**🔥 Official Demo**: [Baidu AI Studio](https://aistudio.baidu.com/application/detail/98365) | **🔥 Official Demo**: [Baidu AI Studio](https://aistudio.baidu.com/application/detail/98365) |
**📝 Blog**: [Technical Report](https://ernie.baidu.com/blog/publication/PaddleOCR-VL_Technical_Report.pdf) **📝 arXiv**: [Technical Report](https://arxiv.org/pdf/2510.14528)
</div> </div>
@ -84,14 +82,18 @@ Install [PaddlePaddle](https://www.paddlepaddle.org.cn/install/quick) and [Paddl
```bash ```bash
python -m pip install paddlepaddle-gpu==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu126/ python -m pip install paddlepaddle-gpu==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu126/
python -m pip install -U "paddleocr[doc-parser]" python -m pip install -U "paddleocr[doc-parser]"
python -m pip install https://paddle-whl.bj.bcebos.com/nightly/cu126/safetensors/safetensors-0.6.2.dev0-cp38-abi3-linux_x86_64.whl
``` ```
> For Windows users, please use WSL or a Docker container.
### Basic Usage ### Basic Usage
CLI usage: CLI usage:
```bash ```bash
paddleocr doc_parser -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pp_ocr_vl_demo.png paddleocr doc_parser -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/paddleocr_vl_demo.png
``` ```
Python API usage: Python API usage:
@ -100,7 +102,7 @@ Python API usage:
from paddleocr import PaddleOCRVL from paddleocr import PaddleOCRVL
pipeline = PaddleOCRVL() pipeline = PaddleOCRVL()
output = pipeline.predict("https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pp_ocr_vl_demo.png") output = pipeline.predict("https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/paddleocr_vl_demo.png")
for res in output: for res in output:
res.print() res.print()
res.save_to_json(save_path="output") res.save_to_json(save_path="output")
@ -117,23 +119,22 @@ for res in output:
--gpus all \ --gpus all \
--network host \ --network host \
ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddlex-genai-vllm-server ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddlex-genai-vllm-server
# You can also use ccr-2vdh3abv-pub.cnc.bj.baidubce.com/paddlepaddle/paddlex-genai-vllm-server for the SGLang server
``` ```
2. Call the PaddleOCR CLI or Python API: 2. Call the PaddleOCR CLI or Python API:
```bash ```bash
paddleocr doc_parser \ paddleocr doc_parser \
-i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pp_ocr_vl_demo.png \ -i https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/paddleocr_vl_demo.png \
--vl_rec_backend vllm-server \ --vl_rec_backend vllm-server \
--vl_rec_server_url http://127.0.0.1:8080 --vl_rec_server_url http://127.0.0.1:8080/v1
``` ```
```python ```python
from paddleocr import PaddleOCRVL from paddleocr import PaddleOCRVL
pipeline = PaddleOCRVL(vl_rec_backend="vllm-server", vl_rec_server_url="http://127.0.0.1:8080") pipeline = PaddleOCRVL(vl_rec_backend="vllm-server", vl_rec_server_url="http://127.0.0.1:8080/v1")
output = pipeline.predict("https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/pp_ocr_vl_demo.png") output = pipeline.predict("https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/paddleocr_vl_demo.png")
for res in output: for res in output:
res.print() res.print()
res.save_to_json(save_path="output") res.save_to_json(save_path="output")
@ -246,31 +247,32 @@ The evaluation set is broadly categorized into 11 chart categories, including ba
### Text ### Text
<div align="center"> <div align="center">
<img src="./imgs/text_english_arabic.jpg" width="300"/> <img src="./imgs/text_english_arabic.jpg" width="300" style="display: inline-block;"/>
<img src="./imgs/text_handwriting_02.jpg" width="300"/> <img src="./imgs/text_handwriting_02.jpg" width="300" style="display: inline-block;"/>
</div> </div>
### Table ### Table
<div align="center"> <div align="center">
<img src="./imgs/table_01.jpg" width="300"/> <img src="./imgs/table_01.jpg" width="300" style="display: inline-block;"/>
<img src="./imgs/table_02.jpg" width="300"/> <img src="./imgs/table_02.jpg" width="300" style="display: inline-block;"/>
</div> </div>
### Formula ### Formula
<div align="center"> <div align="center">
<img src="./imgs/formula_EN.jpg" width="300"/> <img src="./imgs/formula_EN.jpg" width="300" style="display: inline-block;"/>
<img src="./imgs/formula_EN.jpg" width="300"/> <img src="./imgs/formula_ZH.jpg" width="300" style="display: inline-block;"/>
</div> </div>
### Chart ### Chart
<div align="center"> <div align="center">
<img src="./imgs/chart_01.jpg" width="300"/> <img src="./imgs/chart_01.jpg" width="300" style="display: inline-block;"/>
<img src="./imgs/chart_02.jpg" width="300"/> <img src="./imgs/chart_02.jpg" width="300" style="display: inline-block;"/>
</div> </div>
@ -291,4 +293,3 @@ If you find PaddleOCR-VL helpful, feel free to give us a star and citation.
howpublished={\url{https://ernie.baidu.com/blog/publication/PaddleOCR-VL_Technical_Report.pdf}} howpublished={\url{https://ernie.baidu.com/blog/publication/PaddleOCR-VL_Technical_Report.pdf}}
} }
``` ```