26 lines
650 B
Docker
26 lines
650 B
Docker
# Define the image argument and provide a default value
|
|
ARG IMAGE=ghcr.io/abetlen/llama-cpp-python:latest
|
|
|
|
# Define the model file name and download url
|
|
ARG MODEL_FILE=llama-2-7b-chat.bin
|
|
ARG MODEL_DOWNLOAD_URL=https://huggingface.co/TheBloke/Nous-Hermes-Llama-2-7B-GGML/resolve/main/nous-hermes-llama-2-7b.ggmlv3.q4_0.bin
|
|
|
|
FROM ${IMAGE}
|
|
|
|
ARG MODEL_FILE
|
|
ARG MODEL_DOWNLOAD_URL
|
|
|
|
# Download the model file
|
|
RUN apt-get update -y && \
|
|
apt-get install --yes curl && \
|
|
mkdir -p /models && \
|
|
curl -L -o /models/${MODEL_FILE} ${MODEL_DOWNLOAD_URL}
|
|
|
|
WORKDIR /app
|
|
|
|
COPY . .
|
|
|
|
EXPOSE 8000
|
|
|
|
# Run the server start script
|
|
CMD ["/bin/sh", "/app/run.sh"] |