Multi-Model-OCR / requirements.txt
IFMedTechdemo's picture
Update requirements.txt
ccbebae verified
raw
history blame
773 Bytes
# Flash Attention - CUDA 12, PyTorch 2.6, Python 3.10
flash-attn @ https://github.com/Dao-AILab/flash-attention/releases/download/v2.7.3/flash_attn-2.7.3+cu12torch2.6cxx11abiFALSE-cp310-cp310-linux_x86_64.whl
# Core ML/AI Libraries
torch==2.6.0
torchvision
transformers==4.57.1
transformers-stream-generator
accelerate
xformers
# Hugging Face
huggingface_hub
hf_xet
spaces
# Vision & Image Processing
qwen-vl-utils
albumentations
opencv-python
pyvips
pyvips-binary
pillow
timm
einops
supervision
# Document Processing
docling-core
python-docx
pymupdf
pdf2image
markdown
html2text
# PDF Generation
reportlab
fpdf
# Text Processing
sentencepiece
num2words
# Utilities
loguru
requests
httpx
click
# Web Interface
gradio
# Model Fine-tuning
peft
# Video Processing
av