| # Use an alias for the base image for easier updates | |
| FROM python:3.10 as base | |
| # Set model | |
| ENV MODEL=TheBloke/NeuralHermes-2.5-Mistral-7B-GGUF | |
| ENV QUANT=Q4_K_M | |
| ENV CHAT_TEMPLATE=chatml | |
| # Set the working directory | |
| WORKDIR /app | |
| # Install Python requirements | |
| COPY ./requirements.txt /app/ | |
| RUN pip install --no-cache-dir --upgrade -r requirements.txt | |
| # Download model | |
| RUN MODEL_NAME_FILE=$(echo ${MODEL#*/} | tr '[:upper:]' '[:lower:]' | sed 's/-gguf$//') && \ | |
| wget https://huggingface.co/${MODEL}/resolve/main/${MODEL_NAME_FILE}.${QUANT}.gguf -O model.gguf | |
| # Copy the rest of your application | |
| COPY . . | |
| # Command to run the application | |
| CMD ["python", "app.py"] | |