### LLaMACpp Multi-Instance Container with Nginx Load Balancer
### Based on llama-throughput-lab for maximum throughput
### Multiple llama-server instances + nginx for load balancing
###
### BUILD: podman build -t llamacpp:vulkan-multi-amd64 -f llamacpp-multi.Containerfile .
### Export: podman save -o /home/badstorm/llamacpp-vulkan-multi-amd64.tar localhost/llamacpp:vulkan-multi-amd64


FROM ubuntu:24.04

USER root
EXPOSE 8090 9000 9001 9002 9003

RUN apt-get update \
    && apt-get install -y curl tar grep sed git ffmpeg nano python3-pip python3 python3-wheel nginx supervisor \
    && pip install --break-system-packages --upgrade setuptools \
    && pip install --break-system-packages -U "huggingface_hub[cli]" \
    && if [ -f requirements.txt ]; then pip install --break-system-packages -r requirements.txt; fi \
    && apt autoremove -y \
    && apt clean -y \
    && rm -rf /tmp/* /var/tmp/* \
    && rm -rf /var/lib/apt/lists/* \
    && find /var/cache/apt/archives /var/lib/apt/lists -not -name lock -type f -delete \
    && find /var/cache -type f -delete

WORKDIR /app

RUN VERSION=$(curl -s https://api.github.com/repos/ggml-org/llama.cpp/releases/latest | grep '"tag_name"' | head -1 | sed 's/.*"tag_name": "\([^"]*\)".*/\1/') \
    && echo "Last llama.cpp version: $VERSION" \
    && curl -L https://github.com/ggml-org/llama.cpp/releases/download/${VERSION}/llama-${VERSION}-bin-ubuntu-vulkan-x64.tar.gz -o llama.tar.gz \
    && tar -xzf llama.tar.gz -C . --strip-components=1 \
    && rm llama.tar.gz

RUN chmod +x /app/llama-server

# Copy startup script for multiple instances
COPY start-multi-servers.sh /app/bin/
RUN chmod +x /app/bin/start-multi-servers.sh

# Copy nginx config
COPY llama-upstream.conf /etc/nginx/conf.d/

# Copy supervisor config
COPY llama-multi.conf /etc/supervisor/conf.d/

WORKDIR /app

ENV PATH=/app:/app/bin:$PATH
ENV LD_LIBRARY_PATH=/app:$LD_LIBRARY_PATH
ENV HF_HUB_ENABLE_HF_TRANSFER=1
ENV LLAMA_INSTANCES=4
ENV LLAMA_BASE_PORT=9000
ENV LLAMA_ARG_PARALLEL=32
ENV LLAMA_ARG_THREADS=16
ENV LLAMA_ARG_BATCH_SIZE=2048
ENV LLAMA_ARG_CTX_SIZE=131072
ENV LLAMA_ARG_HF_REPO=unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q2_K
ENV LLAMA_ARG_HOST=0.0.0.0
ENV LLAMA_READY_TIMEOUT=600

ENTRYPOINT ["/usr/bin/supervisord"]
CMD ["-c", "/etc/supervisor/conf.d/llama-multi.conf"]