pandas>=2.1
Pillow
pydantic>=2.6
click>=8.1.7
dacite>=1.8
validators>=0.28
python-dotenv>=1.0
typing_extensions<5.0,>=4.15.0
PyYAML>=6.0
setuptools<81

[:python_version < "3.12"]
numpy<2,>=1.26

[:python_version >= "3.12"]
numpy>=2.0

[all]
mmore[api,process,rag]

[api]
fastapi[standard]>=0.110
uvicorn>=0.29
starlette>=0.36
httpx>=0.27
requests>=2.31
pymongo>=4.6
motor>=3.5

[cpu]
torch>=2.7.0
torchvision

[cu126]
torch>=2.7.0
torchvision

[dev]
pytest>=8.3.4
ruff>=0.4.0
pyright

[index]
pymilvus[milvus-lite]==2.6.6
pymilvus-model>=0.3.2
milvus-model>=0.2.12
langchain-milvus>=0.1.8
sentence-transformers
transformers>=4.44
scipy>=1.8

[process]
transformers>=4.44
PyMuPDF
marker-pdf>=1.6
surya-ocr>=0.8.3
moviepy>=2.0
mammoth>=1.8
markdownify>=0.12
markdown>=3.5
python-docx
python-pptx
openpyxl>=3.1
requests>=2.31
trafilatura>=1.12
clean-text
Unidecode
chonkie<1,>=0.2.1
langdetect>=1.0.9
argostranslate
langid
dask[distributed]>=2025.2.0
docx2pdf
lxml_html_clean
beautifulsoup4>=4.12
xlrd>=2.0.1
py7zr>=0.22
rarfile>=4.1
fasteners>=0.19
google-auth>=2.28
google-api-python-client>=2.120
colpali-engine>=0.3
bokeh

[process:python_version < "3.12"]
datatrove>=0.3

[process:python_version >= "3.12"]
datatrove>=0.7

[rag]
mmore[index]
langchain>=0.3
langchain-anthropic>=0.3
langchain-aws>=0.2
langchain-cohere>=0.3
langchain-community>=0.3
langchain-huggingface>=0.1
langchain-mistralai>=0.2
langchain-openai>=0.3
cohere>=5.0
ragas>=0.2
datasets>=4.0
accelerate>=0.30
nltk>=3.9

[websearch]
tavily-python>=0.3.0
ddgs>=6.0
