multimodal-rag-system/requirements.txt at main · SajidAli8015/multimodal-rag-system · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
# Core Dependencies
python-dotenv==1.2.1
pydantic==2.7.4
pydantic-settings==2.12.0


# Document Processing
# Using Docling for unified document parsing (supports PDF, DOCX, PPTX, HTML, images, tables)
docling==2.72.0  # Unified document parser with structure preservation
pymupdf>=1.23.0  # For direct PDF image extraction (fallback when Docling fails)
reportlab==4.4.9  # For PDF generation (testing)
# Legacy parsers (kept as fallback)
unstructured==0.11.6
unstructured-inference==0.7.19
pdf2image==1.16.3
pdfminer.six==20221105
opencv-python==4.8.1.78
matplotlib==3.7.4
pikepdf==10.3.0
python-magic==0.4.27  # For file type detection (macOS)
python-docx==1.2.0  # For DOCX image extraction
python-pptx>=1.0.2,<2.0.0  # For PPTX image extraction
pypdf==3.17.4
pandas==2.1.4
openpyxl==3.1.5  # Note: Upgraded by Docling
tabulate==0.9.0
pillow==10.1.0

# Arabic Text Processing
arabic-reshaper==3.0.0
python-bidi==0.4.2
camel-tools==1.5.2

# Embeddings & Vector Store
sentence-transformers==3.3.1
chromadb==0.4.22
faiss-cpu==1.7.4

# LLM Integration
google-generativeai==0.3.2

# Utilities
langchain==0.1.0
langchain-community==0.0.10
tiktoken==0.5.2
numpy==1.26.3
tqdm==4.66.1

# Web & API
fastapi==0.108.0
uvicorn[standard]==0.25.0
python-multipart>=0.0.9  # For file uploads


# Frontend
gradio==4.44.0
websockets>=10.4,<13.0  # For WebSocket streaming in frontend

# Development
pytest==7.4.3
pytest-asyncio==0.21.1
black==23.12.1
ruff>=0.2.2