Dockerfile

# Use an official Python runtime as a base image
FROM python:3.9-slim

# Set the working directory in the container to /app
WORKDIR /app

# Copy the current directory contents into the container at /app
ADD . /app

# Install any needed packages specified in requirements.txt
RUN apt-get update && \
    apt-get install -y aria2 curl gcc python3-dev && \
    curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y

ENV PATH="/root/.cargo/bin:${PATH}"

RUN pip install --upgrade pip && \
    pip install --no-cache-dir -r requirements.txt

# Install accelerate and bitsandbytes
RUN pip install accelerate
RUN pip install -i https://test.pypi.org/simple/ bitsandbytes

# Create directory and download model
RUN mkdir -p /app/medllama2_7b && \
    chmod -R 777 /app/medllama2_7b && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/resolve/main/model-00001-of-00002.safetensors -d /app/medllama2_7b -o model-00001-of-00002.safetensors && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/resolve/main/model-00002-of-00002.safetensors -d /app/medllama2_7b -o model-00002-of-00002.safetensors && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/raw/main/model.safetensors.index.json -d /app/medllama2_7b -o model.safetensors.index.json && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/raw/main/special_tokens_map.json -d /app/medllama2_7b -o special_tokens_map.json && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/raw/main/tokenizer.json -d /app/medllama2_7b -o tokenizer.json && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/raw/main/tokenizer_config.json -d /app/medllama2_7b -o tokenizer_config.json && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/raw/main/config.json -d /app/medllama2_7b -o config.json && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/raw/main/generation_config.json -d /app/medllama2_7b -o generation_config.json && \
    aria2c --console-log-level=error -c -x 16 -s 16 -k 1M https://huggingface.co/4bit/medllama2_7b_s/resolve/main/tokenizer.model -d /app/medllama2_7b -o tokenizer.model

# Make port 80 available to the world outside this container
EXPOSE 80

# Run app.py when the container launches
CMD ["python", "app.py"]