requirements-base.txt

# ============================================
# OpenNotebookLM - Python Dependencies
# ============================================

# ------ Core Framework ------
fastapi==0.135.1
uvicorn==0.29.0
starlette==0.52.1
pydantic>=2.11.0
pydantic-settings>=2.12.0
typing-extensions==4.15.0
python-dotenv==1.0.1
python-multipart>=0.0.22
httpx[socks]==0.28.1
requests==2.32.5
charset-normalizer==3.4.4
supabase==2.27.2

# ------ LLM / LangChain ------
openai==2.16.0
aisuite>=0.1.14
langchain==0.3.27
langchain-chroma==0.2.6
langchain-community==0.3.29
langchain-core==0.3.76
langchain-openai==0.3.33
langchain-text-splitters==0.3.11
langgraph==0.6.7
langgraph-checkpoint==2.1.1
langgraph-prebuilt==0.6.4
langgraph-sdk==0.2.6
sseclient-py>=1.8.0

# ------ Vector Store / Embedding ------
chromadb==1.5.5
faiss-cpu==1.12.0
sentence-transformers>=3.0.0
onnxruntime==1.23.2
huggingface-hub==0.36.2

# ------ Document Parsing ------
PyMuPDF==1.26.7
pdfplumber==0.11.9
pdfminer.six==20251230
PyPDF2>=3.0.1
pdf2image>=1.17.0
trafilatura>=2.0.0
chonkie>=1.3.0
lxml_html_clean>=0.4.0

# ------ Office / Export ------
python-pptx>=1.0.2
python-docx>=1.2.0
openpyxl>=3.1.5
xlsxwriter==3.2.9

# ------ Image / Vision ------
pillow>=10.4.0
opencv-python>=4.11.0
cairosvg>=2.8.0
imgkit>=1.2.3
vtracer>=0.6.0
matplotlib>=3.10.0
scikit-image==0.26.0
ultralytics==8.3.241

# ------ ML / Deep Learning ------
numpy<2.0.0
pandas>=2.3.0
scipy>=1.16.0
sqlalchemy==2.0.25
sqlmodel==0.0.37
duckdb==1.5.0
torch
torchvision
torchaudio
transformers==4.57.3
accelerate>=1.12.0
scikit-learn>=1.8.0

# ------ Web Scraping / Search ------
beautifulsoup4>=4.14.0
playwright==1.56.0

# ------ Deep Research (Qwen-Agent) ------
qwen-agent>=0.0.34
alibabacloud-docmind-api20220711>=1.4.13
alibabacloud-credentials==1.0.7
alibabacloud-tea-openapi==0.4.3
alibabacloud-tea-util==0.3.14
sandbox-fusion>=0.3.7
tiktoken>=0.12.0
tabulate>=0.9.0
json5>=0.12.0

# ------ Workflow Engine (DataFlow-Agent) ------
open-dataflow>=1.0.8
cloudpickle>=3.1.0
psutil>=7.2.0
pyfiglet>=1.0.4
pyyaml>=6.0.0
termcolor>=3.2.0
colorlog>=6.10.0
colorama>=0.4.6
rich>=14.0.0
tqdm>=4.67.0
jinja2>=3.1.0

# ------ Data Quality / NLP ------
datasets<=3.2
evaluate>=0.4.6
nltk>=3.9.0
sentencepiece>=0.2.1
rapidfuzz>=3.14.0
datasketch>=1.8.0
simhash>=2.1.0
contractions>=0.1.73
symspellpy>=6.9.0

# ------ Text Quality / PII ------
fasttext-wheel>=0.9.2
langkit>=0.0.35
presidio_analyzer[transformers]>=2.2.0
presidio_anonymizer>=2.2.0
vendi-score==0.0.3
google-api-core>=2.28.0
google-api-python-client>=2.187.0

# ------ Audio / TTS ------
librosa>=0.11.0
soundfile>=0.13.0
qwen-tts==0.1.1
fireredtts2==0.1

# ------ OCR (PaddlePaddle / Local Serving) ------
paddlepaddle==3.2.0
paddleocr==2.9.1
mineru_vl_utils==0.1.21
vllm==0.16.0
vllm-omni==0.16.0
loguru>=0.7.3

# ------ Optional Feature Extras (not installed in current szl-dev) ------
# jieba>=0.42.1            # Better Chinese tokenization for SQLBot RAG
# rank-bm25>=0.2.2         # BM25 retrieval for schema/value linking
# elasticsearch>=8.0.0     # Native Elasticsearch datasource support
# oracledb>=2.0.0          # Oracle datasource support
# scenedetect>=0.6.0       # Better video keyframe extraction
# ffmpeg-python>=0.2.0     # Python FFmpeg wrapper for media analysis

# ------ Misc ------
appdirs>=1.4.4
addict>=2.4.0
docstring_parser>=0.15
math_verify>=0.8.0
word2number>=1.1
func_timeout>=4.3.5
sqlglot>=28.0.0
pymysql>=1.1.2
modelscope>=1.33.0
gradio>=5.49.0
IPython==8.18.1

# ------ Dev / Test ------
pytest>=9.0.0