https://github.com/PaddlePaddle/PaddleOCR

1. 创建虚拟环境

# Python 3.8.10
python3 -m venv paddle_env
source paddle_env/bin/activate

2. 安装依赖

pip install --upgrade pip
pip install paddlepaddle-gpu==2.5.2 -f https://www.paddlepaddle.org.cn/whl/linux/mkl/avx/stable.html

3. 安装 PaddleOCR

pip install paddleocr==2.7.3

4. 验证

python -c "from paddleocr import PaddleOCR; print('OK')"

5. pip包

pip list
Package                        Version
------------------------------ -----------
albucore                       0.0.17
albumentations                 1.4.18
annotated-types                0.7.0
anyio                          4.5.2
astor                          0.8.1
attrdict                       2.0.1
babel                          2.18.0
bce-python-sdk                 0.9.72
bcrypt                         5.0.0
beautifulsoup4                 4.15.0
cachetools                     5.5.2
certifi                        2026.6.17
cffi                           1.17.1
charset-normalizer             3.4.9
ci-info                        0.4.0
click                          8.1.8
coloredlogs                    15.0.1
contourpy                      1.1.1
crc32c                         2.8
cryptography                   47.0.0
cssselect                      1.2.0
cssutils                       2.11.1
cycler                         0.12.1
Cython                         3.2.8
decorator                      5.3.1
dill                           0.4.0
et_xmlfile                     2.0.0
etelemetry                     0.3.1
eval_type_backport             0.4.0
exceptiongroup                 1.3.1
filelock                       3.16.1
fire                           0.7.1
Flask                          2.0.2
Flask-Babel                    2.0.0
Flask-Cors                     3.0.10
flatbuffers                    25.12.19
fonttools                      4.57.0
funcsigs                       1.0.2
future                         1.0.0
gast                           0.3.3
graphviz                       0.20.3
h11                            0.16.0
httpcore                       1.0.9
httpx                          0.28.1
humanfriendly                  10.0
idna                           3.15
imageio                        2.35.1
imgaug                         0.4.0
importlib_resources            6.4.5
isodate                        0.6.1
itsdangerous                   2.2.0
Jinja2                         3.1.6
joblib                         1.4.2
kiwisolver                     1.4.7
lazy_loader                    0.4
Levenshtein                    0.25.1
lmdb                           1.4.1
looseversion                   1.3.0
lxml                           6.1.1
MarkupSafe                     2.1.5
matplotlib                     3.6.3
more-itertools                 10.5.0
mpmath                         1.3.0
multiprocess                   0.70.18
networkx                       3.1
nibabel                        5.1.0
nipype                         1.8.6
nltk                           3.9.1
nodeenv                        1.6.0
numpy                          1.24.4
objgraph                       3.6.2
onnxruntime                    1.19.2
opencv-contrib-python          4.4.0.46
opencv-contrib-python-headless 4.2.0.32
opencv-python                  4.2.0.32
opencv-python-headless         5.0.0.93
openpyxl                       3.0.10
opt-einsum                     3.3.0
packaging                      21.3
paddle-bfloat                  0.1.2
paddleocr                      2.8.1
paddlepaddle                   2.6.2
paddlepaddle-gpu               2.5.2
pandas                         1.3.3
paramiko                       2.8.0
pathlib                        1.0.1
pdf2docx                       0.5.8
pdf2image                      1.16.3
pillow                         10.4.0
pip                            25.0.1
premailer                      3.10.0
prettytable                    3.11.0
protobuf                       3.20.2
prov                           2.0.1
pyclipper                      1.3.0.post6
pycparser                      2.23
pycryptodome                   3.23.0
pydantic                       2.10.6
pydantic_core                  2.27.2
pydot                          4.0.1
PyMuPDF                        1.23.3
PyMuPDFb                       1.23.3
PyNaCl                         1.6.2
pyOpenSSL                      19.1.0
pyparsing                      3.1.4
PyPDF2                         3.0.1
python-dateutil                2.9.0.post0
python-docx                    1.1.2
python-Levenshtein             0.25.1
pytz                           2026.2
PyWavelets                     1.4.1
PyYAML                         6.0.3
rapidfuzz                      3.9.7
rarfile                        4.2
rdflib                         6.3.2
regex                          2024.11.6
requests                       2.32.4
scikit-image                   0.21.0
scipy                          1.10.1
setuptools                     75.3.4
shapely                        2.0.7
simplejson                     4.1.1
six                            1.17.0
sniffio                        1.3.1
soupsieve                      2.7
sympy                          1.13.3
termcolor                      2.4.0
tifffile                       2023.7.10
tqdm                           4.68.4
traits                         6.3.2
typing_extensions              4.13.2
urllib3                        2.2.3
visualdl                       2.4.2
wcwidth                        0.8.2
Werkzeug                       2.0.2
wheel                          0.45.1
zipp                           3.20.2

Windows系统下部署

1. 先升级基础工具

pip install --upgrade pip setuptools wheel

2. 安装 patch-ng(解决 lmdb 编译问题)

pip install patch-ng

3. 安装 paddlepaddle(新版兼容 Python 3.8)

pip install paddlepaddle==2.6.2

尝试安装有预编译 wheel 的版本

pip install lmdb==1.4.1
pip install attrdict cython premailer
pip install pdf2docx python-docx