PaddleOCR-VL GGUF for llama.cpp
Tested with Win10 x64 CPU only
With command:
llama-server -m D:\mantownenlllr1\PaddleOCR-VL-GGUF.gguf --mmproj D:\mantownenlllr1\PaddleOCR-VL-GGUF-mmproj.gguf --chat-template-file D:\mantownenlllr1\PaddleOCR-VL-Chat.jinja --cache-type-k q8_0 --cache-type-v q8_0 --port 8900 --temp 0 --top-k 0 --top-p 0 --min-p 0 --repeat-last-n 0 --repeat-penalty 0
