OCR / requirements.txt
pankti07's picture
Update requirements.txt
c0eef89 verified
raw
history blame contribute delete
447 Bytes
--extra-index-url https://download.pytorch.org/whl/cpu
torch>=2.4.0
torchvision>=0.19.0
# CRITICAL: Qwen2.5-VL is only supported in the development version of transformers right now
git+https://github.com/huggingface/transformers.git
accelerate>=0.26.0
gradio>=4.0.0
pdf2image
pillow
einops
# Required helper for Qwen-VL models
qwen-vl-utils
# Optional: For loading simple bit-quantization if needed (often GPU only, but good to have)
bitsandbytes