Spaces:
Running on Zero
Running on Zero
run locally on windows using llama.cpp howto
#16
by douchebanner - opened
based on information from previous discussions, namely https://huggingface.co/spaces/fancyfeast/joy-caption-beta-one/discussions/7 and https://huggingface.co/spaces/fancyfeast/joy-caption-beta-one/discussions/9
i assume you are familiar with llama.cpp and that you already have gemma or other models running on llama.cpp
open a command line on your llama.cpp folder and run
llama-cli --hf-repo concedo/llama-joycaption-beta-one-hf-llava-mmproj-gguf --hf-file Llama-Joycaption-Beta-One-Hf-Llava-Q8_0.gguf
then run the model as you normally would. i run the server and then use the web gui to load the model and drag and drop the images.
if i had known it was this easy i would have done it sooner XD