Dockerfile

# Build stage
FROM golang:1.24-alpine AS builder

WORKDIR /app

# Copy go mod files
COPY go.mod go.sum ./
RUN go mod download

# Copy source files
COPY . .

# Build
RUN CGO_ENABLED=0 GOOS=linux go build -o /app/main

# Final stage
FROM ubuntu:22.04

# Copy built binary from builder
COPY --from=builder /app/main /app/main
RUN apt-get update && \
    apt-get install -y \
    build-essential \
    libssl-dev \
    zlib1g-dev \
    libboost-system-dev \
    libboost-filesystem-dev \
    cmake \
    git \
    python3-pip \
    curl \
    wget && \
    rm -rf /var/lib/apt/lists/*

# Install Python dependencies
RUN pip3 install huggingface-hub openai gradio

# Build llama.cpp
RUN git clone https://github.com/ggerganov/llama.cpp && \
    cd llama.cpp && \
    mkdir build && \
    cd build && \
    cmake .. -DLLAMA_BUILD_SERVER=ON -DLLAMA_BUILD_EXAMPLES=ON -DCMAKE_BUILD_TYPE=Release && \
    cmake --build . --config Release --target llama-server -j $(nproc)

# Download model
RUN mkdir -p /models && \
    wget -O /models/model.q8_0.gguf https://huggingface.co/unsloth/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/resolve/main/DeepSeek-R1-Distill-Qwen-1.5B-Q8_0.gguf

# Copy startup script
COPY start.sh /start.sh
RUN chmod +x /start.sh

# Expose ports
EXPOSE 8080 3000

# Start services
CMD ["/start.sh"]