Make GPU parameters optional and easier to manage

Dockerfile

@@ -1,12 +1,12 @@
 FROM nvidia/cuda:11.8.0-devel-ubuntu22.04
 ARG MODEL
 ARG MODEL_NAME
-ARG CONTEXT_SIZE
+ARG ADDITIONAL
 RUN mkdir /opt/koboldcpp
 RUN apt update && apt install git build-essential libopenblas-dev wget python3-pip -y
 RUN git clone https://github.com/lostruins/koboldcpp /opt/koboldcpp
 WORKDIR /opt/koboldcpp
 RUN make LLAMA_OPENBLAS=1 LLAMA_CUBLAS=1 LLAMA_PORTABLE=1
 RUN wget -O model.ggml $MODEL
-CMD /bin/python3 ./koboldcpp.py --model model.ggml --usecublas --gpulayers 99 --multiuser --contextsize $CONTEXT_SIZE --port 7860 --hordeconfig $MODEL_NAME 1 1
+CMD /bin/python3 ./koboldcpp.py --model model.ggml $ADDITIONAL --port 7860 --hordeconfig $MODEL_NAME 1 1
 

README.md

@@ -24,7 +24,5 @@ This space was designed to be easy to clone, first make sure you convert your mo
 
 If you have a GPU available for your space, clone this space and point the MODEL variable to your model's download location, then force a rebuild so it can use your own custom model. You can customize the model that is being displayed by setting the MODEL_NAME.
 
-Want to run on the CPU tier? The following line enables multiuser GPU usage.
-, "--usecublas", "mmq", "--gpulayers", "99", "--multiuser", "--contextsize", "4096"
-If you remove this from the CMD in the Dockerfile your instance will now be compatible with CPU only usage.
+In additional variable we configured the parameters required to run this on a GPU and support multiple users and high context, if you wish to clone this to a CPU space simply leave that blank.